|
 |
Iversen’s Multiconfused Log (see p.1!) | ||
Tags: Multilingual | ||
Share with:
| ||
 Language Learning Forum : Language Learning Log |  | |
 Iversen Super Polyglot Moderator Denmark berejst.dk Joined 6492 days ago 9078 posts - 16473 votes  Speaks: Danish*, French, English, German, Italian, Spanish, Portuguese, Dutch, Swedish, Esperanto, Romanian, Catalan Studies: Afrikaans, Greek, Norwegian, Russian, Serbian, Icelandic, Latin, Irish, Lowland Scots, Indonesian, Polish, Croatian Personal Language Map | Message 3737 of 3959 04 November 2014 at 12:55pm | IP Logged |
|
In my intensive studies of Bahasa Indonesia I have for some time used a very long and quite good text about the history of the Earth and its life forms. I intend to continue using this text, but the section I read yesterday had some elements that suggest that the article may be translated from English - and translated via a machine translation. The details follow (in Indonesian).
ESP: Enbusiĝinta hejmen de laboro, mi hieraŭ legis interŝanĝon de opinioj pri la problemoj de akiri literaturon en Esperanto, kiam ordinara librejoj ne vendus librojn en Esperanto kaj oni povas nur esti akirita per interreto. Ĝi aperas ke la nombro da titoloj kreskas, sed la nombro de legantoj falis. SW: Og vidrörande tv: jag tittat nog mest engelskspråkiga program, men igår såg jag et populärvetenskaplig show från Sverige, som handlade om så vitt skilda ämnen som återuppbyggnaden av tarmfloran med riktig människobajs och sonden Rosettas studier av is-kometen Gerasimov. Gerasimov liknar på en jordnöt - eller sagt kanske en gummianka, som de sa programmet. Denna sond kretsar redan kring kometen, men kommer att försöka skicka en liten sak kallat Philae ned till ytan av kometen, men det kommer att vara mycket svårt, eftersom den är mycket ujämnare än väntat. Kanske det går bra, kanske det kommer att gå fel. I det senare fallet har dom bortkastat ett par miljoner euro (og kommar at gråta som små barn). BA I: Dalam artikel tentang kisah Bumi kini saya telah membaca bab tentang kehidupan di usia Ordovician. Beberapa deskripsi terlihat aneh. Misalnya ostracoderms digambarkan sebagai "tanpa rahang, ikan lapis baja". Mengapa tidak "ikan lapis baja tanpa rahang"'? Kemudian dalam bab ini Inggris kata kata "sea", "platelike", "slitlike" dan "anterior" digunakan. Yang membuat saya curiga terjemahan mesin, yang setidaknya harus dikoreksi oleh manusia sebelum publikasi. Sampai bab ini, saya tidak curiga apa-apa permainan kotor, dan saya akan terus bekerja dengan itu karena sangat menarik dan saya belajar banyak dari kata-kata lucu - tapi itu mengkhawatirkan, bahwa beberapa bagian telah diterjemahkan oleh Google. Bagaimana sisanya? PS: an ostracoderm is (or rather was) a 'jawless, plate armoured fish'. They died out out about 420 mio. years ago, and their place in the ecosystem was taken by sharks and other more advanced creatures. Edited by Iversen on 04 November 2014 at 2:41pm 1 person has voted this message useful | |
| Iversen Super Polyglot Moderator Denmark berejst.dk Joined 6492 days ago 9078 posts - 16473 votes Speaks: Danish*, French, English, German, Italian, Spanish, Portuguese, Dutch, Swedish, Esperanto, Romanian, Catalan Studies: Afrikaans, Greek, Norwegian, Russian, Serbian, Icelandic, Latin, Irish, Lowland Scots, Indonesian, Polish, Croatian Personal Language Map | Message 3738 of 3959 06 November 2014 at 4:33pm | IP Logged |
|
Tuesday evening I made an Italian wordlist with about 100 words because I thought that my Italian needed some fresh input.Then I got the idea that it might be nice to have a quantitative assessment of the need for fresh input and I made a word count of the kind where I divide the words on a number of pages in a dictionary into known, not known and something in between. I used my old Italian-Danish Gyldendal for the purpose and assessed 363 words from 4 pages, and I got the following results: known 53% (an estimated 23000 words), so-so 17% (7000) and not known 30% (13000 words out of a total of 43000 words). With so few pages in the counts there is bound to be some slack in these figures, but compared with my numbers from 2013 the sum of known and so-so (or 'guessable') still lie around 70% of the words in a midsized dictionary with some 40.000 words. I have just used the middle category more this time, after I stopped calling it 'guessable'.
Wednesday evening I got the idea that it would be be nice also to have estimates for other languages, and I proceeded to make such estimates based on around 200 words for the other Romance languages (3-4 pages in each dictionary) which isn't much, but I wanted to get through all the Romance languages fast). The results were more or less as expected with one notable exception: Romanian. First I counted the first two columns on just 3 pages from my Teora dictionary, which gave 228 words. But all these pages happened to hit upon areas with many international words, so I got a whopping 70% known words, which is equal to some 32000 words out of 46000. I found that result totally unrealistic so I added a couple of pages more, which brought the result down to 62% and 29000 known words – still way too high, but at least the last page I went through demonstrated the other part of the spectrum: this page was dominated by 'native' words (from "zdrobit" to "zgâria") and here I only knew a third of the words. And is this cheating? Yes of course it is (adding more test pages to get a more realistic result is not part of any true scientific ethos), but it illustrates one pitfall of vocabulary assessments in general, namely the tendency to let international loanwords dominate. In this case it is simply by their sheer number, but in the tests on the internet it also seems that people equate a large vocabulary with knowledge of 'learned' words – where my results from Romanian demonstrate that you can have a deficit of native words which is hidden by good results with borrowed vocabulary. Apart from that I also read the beginning of a fictional book, but... FR: Le livre en question était parmi ceux que j'ai acquiert durant mes études romanes dans les années 70 et ne point ouvert depuis lors: "Le Bal du comte d'Orgel" de Raymond Radiguet. Je l'ai lu jusqu'au point où les personnages principaux regardent des clowns dans un cirque. Le couple Orgel décide qu'ils préfèreraient être assis près d'un de leurs connaissances (un certain monsieur Paul Quelquechosequinemintéressepas), et voilà ce qui m'a fait remettre ce livre sur l'étagère: pour faire place au couple arrogant et dégoûtant l'ouvreuse fait bouger des innocents spectateurs de leur places. Et monsieur Radiguet ne fait rien du tout pour punir cette action tout à fait inadmissible, encore qu'il eût pu le faire comme auteur de ce livre. Est-ce qu'il vraiment faut accepter une telle complicité de crimes horribles simplement parce qu'ils se déroule dans un monde fictif? Il n'y a que deux personne dans ce livre que je trouve vaguement symphatiques: la négresse qui n'a pas des devoirs bien précis dans la maison des Orgel et pour cette raison on attends qu'elle fasse TOUT, et un certain hobereau prussien qui est collectionneur: il s'est fait un collection des virgules qui se trouvent dans une édition de Dante (p.51). OK, il est aussi un alcoolique, mais dans une oeuvre littéraire personne n'est parfait, même pas un collectionneur de virgules. Son fils est un monstre hydrocéphalique, ce qui prouve la malveillance systématique de Mr. Radiguet. Edited by Iversen on 06 November 2014 at 4:50pm 1 person has voted this message useful | |
| Iversen Super Polyglot Moderator Denmark berejst.dk Joined 6492 days ago 9078 posts - 16473 votes Speaks: Danish*, French, English, German, Italian, Spanish, Portuguese, Dutch, Swedish, Esperanto, Romanian, Catalan Studies: Afrikaans, Greek, Norwegian, Russian, Serbian, Icelandic, Latin, Irish, Lowland Scots, Indonesian, Polish, Croatian Personal Language Map | Message 3739 of 3959 07 November 2014 at 11:40am | IP Logged |
|
Yesterday I added Latin, Greek Dhimotiki, Esperanto and Indonesian to the wordcount package. With Latin I used two very different dictionaries: my excellent Langenscheidt, where the number of word is limited, but the explanations are extremely detailed with many idiomatic expressions included, and my excellent New College thing from Manila, which is the one I use when I want to write an Latin about modern themes. I know 63% words out of 161 from Langenscheidt and 56% out of 216 from New College - both numbers are higher than those from previous rounds with the same dictionary, where the minimum was 30% in 2013 and the maximum 43% in 2009. The fluctuations can be ascribed to differences between the chosen pages. I use the following method to choose pages: first I open a random place for page no. 1 and then I check equidistant pages to avoid any conscious bias. And sometimes that takes me to a page full of problem words, at other times I get a very easy page full of international terms. That's life.
In Greek I used my Pataki Greek-Danish dictionary (with approx. 27.000 headwords) and got 51% out of 246 – much better than my results with the same dictionary from 2009 (29%) and 2013 (23%). It seems that my Greek wordlists and increased reading activity finally are having some consequences. My largest Esperanto dictionary is an old Teach Yourself thing from the time where dinosaurs roamed the Earth (but still post-Zamenhof). I got 73% in 2013 and 76% now, and with my small samples these numbers are indistinguishable. Or in other words: status quo. Finally I got 36% out of 189 word families with my Tuttle Indonesian dictionary. In 2013 I got 22%, but due to the structure of the dictionary the numbers are small – roughly 3500 word families this time. Why word families when I usually use the head words? Because each article in Tuttle contains a blue headword (sometimes two), some derivations based on affixes and some word combinations of the type we know from English, where each compound has a very specific and often 'unguessable' meaning. To complicate things the derivations also have their own articles, so counting them in the articles would mean that they where included several times. Or in other words: with a dictionary that already is structured according to word families the logical thing is to follow the structure of the book. Roughly speaking my level in the Romance languages is stable at a reasonably high level, and my level in the 'peripheral' languages is higher now than it was just one year ago. That's fine. I'll add word counts for a number of Germanic and Slavic languages in the near future. By the way: in Jyllandsposten today there was a full page (in Danish) written by Knud Andersen (teacher, translator and interpreter of German) who deplores the total lack of respect for other languages than English from the authorities and in the electronic media. It contained a good example of the catastrophic consequences of relying on machine translations or unqualified personnel. I quote: DA: En dansk maskinfabrik havde kontakt på levering af en mobil kran til Tyskland. If. kontrakten skulle leverandøren sørge for "Geländesicherung". Det var fejlagtigt oversat som sikring med et gelænder, hvilket der var en selvfølge og en ringe udgift. Men "Gelände" betyder terræn, og det drejede sig om indhegning af et større område med et solidt trådhegn, hvad der var en kostbar affære. EN: ('gelænder' = a hand rail, 'Gelände' = an area) So because of a translation error the company had to pay a fence around a whole building site, where they had calculated with a cheap handrail somewhere. DA: Vores undervisningsminister og hendes hærskarer af Djøf'er (medlemmer af dansk jurist- og økonomforbund) synes stadig det er en god ide at beskære optaget til sprogfagene på universitetsniveau, fordi en eller anden kommission på et tidspunkt har udtalt en ikke nærmere underbygget frygt for arbejdsløshed på området. End ikke kinesisk går efter planerne fri, og europæiske sprog som tysk og fransk skal åbenbart beskæres til samme niveau som ungarsk havde i min studietid. Og sprogundervisningen er for længst forsvundet fra Danmarks Radio. Som jeg tidligere har skrevet, går det fint med at øge kendskabet til engelsk i Danmark, men alle andre sprog bliver klart nedprioriteret. Edited by Iversen on 10 November 2014 at 2:16pm 3 persons have voted this message useful | |
| tarvos Super Polyglot Winner TAC 2012 Senior Member China likeapolyglot.wordpr Joined 4496 days ago 5310 posts - 9399 votes Speaks: Dutch*, English, Swedish, French, Russian, German, Italian, Norwegian, Mandarin, Romanian, Afrikaans Studies: Greek, Modern Hebrew, Spanish, Portuguese, Czech, Korean, Esperanto, Finnish | Message 3740 of 3959 07 November 2014 at 12:49pm | IP Logged |
|
NL: Hier in Nederland worden helaas ook steeds meer talenstudies samengevoegd tot
grotere studies - verschillende zijn al verloren gegaan, of opgegaan in andere opleidingen. Engels kent nog steeds een hoog aantal studenten, uiteraard, maar je merkt wel de terugloop in kennis, ook voor Frans en Duits, wat ik vreemd vind, want dat zijn en blijven schoolvakken hier. In Amsterdam is er nu een opleiding voor talen in de Balkanregio, waar talen die genetisch maar zijdelings aan elkaar verwant zijn (zoals Roemeens, Servo-Kroatisch, Grieks en Bulgaars) onderwezen als onderdeel van een bachelor Zuidoost-Europese talen. Dit omdat de interesse voor deze talen gestaag is gedaald over de afgelopen jaren. Zelfs talen als Frans en Duits ontkomen hier dus niet aan, alleen het Engels blijft gevrijwaard hiervan. Men vergeet echter dat de kennis van buurtalen juist essentieel is voor het doen van goede handel, en het is ook niet zo dat het Duits nou zo ver af staat van het Nederlands. Dit geldt eigenlijk precies zo voor Denemarken. Ik vind dit een beetje jammer, want ik blijf erbij dat de kennis van enkele vreemde talen buiten het Engels nog steeds een groot átout is voor de meeste mensen, zeker in een commerciële omgeving, en er is ook duidelijk vraag naar. 2 persons have voted this message useful | |
| Iversen Super Polyglot Moderator Denmark berejst.dk Joined 6492 days ago 9078 posts - 16473 votes Speaks: Danish*, French, English, German, Italian, Spanish, Portuguese, Dutch, Swedish, Esperanto, Romanian, Catalan Studies: Afrikaans, Greek, Norwegian, Russian, Serbian, Icelandic, Latin, Irish, Lowland Scots, Indonesian, Polish, Croatian Personal Language Map | Message 3741 of 3959 09 November 2014 at 11:18pm | IP Logged |
|
DU: Tarvos wijst op een groot probleem met de nieuwe curricula, namelijk dat de taalstudies van gebiedstudies opslokken woorden - en wat gebeurt er dan? Ja, de studenten geloven niet meer dat ze nodig hebben om de taalen te leren, maar dat het genoeg is bronteksten over de geschiedenis en cultuur vertaald in het Engels te lezen. En de juristen en economen van het ministerie en hun handlangers buiten ontsnappen niet eens dat er een probleem is.
EN: And speaking about Dutch... I have been making more micro word counts this weekend, where I have visited my mother. Her collection of dictionaries is not quite as extensive as mine, but she has German, English and French dictionaries from Gyldendal and an Oxford Concise dictionary on her shelves so I went for these three languages. And after I had returned home I added Afrikaans and Dutch. In the French dictionary (from 1933!) I got 76%, which is higher than my result from Thursday, but not to an unreasonable degree. Than I did the same job with a somewhat newer German-Danish dictionary and got 75% - well, I have had similar results before. And in English I got 78% with the Concise wonder from 1976, which according to my calculations gives an estimate of 35.000 words out of some 45.000 words. I have used this dictionary before, once where I counted everything and another where I only counted the headwords (defined as the word that initiates each article). This time I counted all single-word items except abbreviations and proper names, but as usual the important thing is not what you count - 78% is 78% no matter what the unit you choose, and the absolute numbers are too reliant on the dictionary size to be of much value. Maybe I should add here that I own two mega-dictionaries: Bratli's Spanish-Danish unequalled monster with at least 200.000 headwords and Webster's unabridged behemoth with some 165.99 words or so. And with both I got lower percentages than with the smaller dictionaries, but still with higher absolute estimated numbers. This can probably be interpreted like a sign that there is a 'true' upper limite for my passive vocabularies somewhat higher than the numbers I get with dictionaries at 30-60.000 words - but not dramatically higher numbers. But then I returned home and got some troubling results. First I counted words in my tiny Sass Low German dictionary and got 70% known words - but out of a mere 5-6000 words so it doesn't mean much. Then I proceeded to count words in my Pharos Afrikaans school dictionary with a respectable 46.000 words or so, and here I got 66% (against 36% in 2009). OK, I have studied the language for some time and I visited South Africa in 2013, but as part of a German/English speaking group. And it has been hard to find reading and listening materials so I can't point to a lot of activity with this language recently. Mystery... Then I did a word count with a thick Gyldendal Dutch-Danish dictionary with around 40.000 headwords - and got 77%! OK, I did a monolingual new years trip ten months ago, but this year I have been occupied by my Slavic languages and Dutch has definitely not been at the forefront of my studies. So to do one more test of the Ingväonian languages I used my old Prisma Dutch-German dictionary with around 29000 words and got 68%. OK, 68 is less than 77, but still unlikely - I speak French better than Dutch and have read and listened much more to French than to Dutch so why do I then get results in the same range? One thing I have noticed is that Dutch and Afrikaans just like German use a lot of compound words, where French uses combinations of free words with or without a preposition. So the cure for impossibly high Dutch/Afrikaans figures might be to count word families and ignore the compound words. But that'll have to wait. The next couple of evenings I'll be counting words in a few Slavic and Scandinavian languages. And maybe Irish, but if I get more than a few words there I'll be extremely surprised. Edited by Iversen on 09 November 2014 at 11:39pm 1 person has voted this message useful | |
| Iversen Super Polyglot Moderator Denmark berejst.dk Joined 6492 days ago 9078 posts - 16473 votes Speaks: Danish*, French, English, German, Italian, Spanish, Portuguese, Dutch, Swedish, Esperanto, Romanian, Catalan Studies: Afrikaans, Greek, Norwegian, Russian, Serbian, Icelandic, Latin, Irish, Lowland Scots, Indonesian, Polish, Croatian Personal Language Map | Message 3742 of 3959 11 November 2014 at 9:56am | IP Logged |
|
Yesterday I worked until late in the evening, so I just had time for two word counts: Icelandic and Swedish. Icelandic landed as expected in the low fifty percents, Swedish in the seventies - I don't have the precise numbers here, but they will be shown here together with the whole busload of results when I have counted the last few languages. But in the meantime there may be a reason to discuss the fluctuating total word numbers for even one and the same dictionary. The clearest example is my use of Oxford Concise Dictionary, which I have used twice in 2013 and one in 2014. I base my estimates of the totals on the words I have counted divided by the pages used, multiplied by the total number of pages with words (excluding grammar pages etc.).
I do this because I found out that the published figures are very unreliable - even when expressions allegedly aren't included. The lowest total for the concise one is 33.000, the newest figure is 45.000, and the highest figure is 74.000 items. And the reason is that there always is something to exclude, like abbreviations and - maybe - proper names, expressions, 'fixed' compunds and examples. So in 2013 I made one count where I only looked at the first word in each article and another where I included just about anything. This time I included single-word compounds, but not those consisting of several wods, because there isn't any objective way to separate those from examples and expressions. On the other hand I have come to the conclusion that proper names which are very different from the native versions in other language should be included - "Leghorn" for Livorno (in Italy) is as much a word as 'city'. With my Swedish investigation yesterday (based on a thick red Gyldendal Swedish-Danish) he happened to count the page with "säng" (bed), and there are scores of compund words with "säng" as the first part - so my 'know' category just grew and grew. So the fluctuating totals for some dictionaries which I have used several times reflect both diffferent weays of counting, but also the issue with 'easy' pages with tons of compounds against pages with very few, 'big' words with a lot of examples which aren't counted. Word counting may actually become an interesting and maybe even fruitful activity when you have done so many counts that you can see some kind of system in the data, but it is far from being an exact science. Edited by Iversen on 11 November 2014 at 12:03pm 1 person has voted this message useful | |
| Iversen Super Polyglot Moderator Denmark berejst.dk Joined 6492 days ago 9078 posts - 16473 votes Speaks: Danish*, French, English, German, Italian, Spanish, Portuguese, Dutch, Swedish, Esperanto, Romanian, Catalan Studies: Afrikaans, Greek, Norwegian, Russian, Serbian, Icelandic, Latin, Irish, Lowland Scots, Indonesian, Polish, Croatian Personal Language Map | Message 3743 of 3959 11 November 2014 at 1:15pm | IP Logged |
|
IT: Quando faccio stampe per studio intensivo cerco spesso di concentrare su qualche tema comune. Ora ho letto la maggior parte delle stampe che feci con temi della storia della Terra, dunque ieri ho fatto una nuova pila di articoli sulla fisica nucleare. E perché il mio Italiano ha bisogno di un poco attenzione e perché il signor Majorana è (o era) Italiano, ho scelto articoli sui fermioni di Majorana (anche Enrico Fermi che ha dato il suo nome ai fermioni era Italiano).
I fermioni hanno massa e devono ubbedire al principio di Pauli (terzo fisico di origine Italiana): non starci nello stezzo luogo che un altro fermione. I bosoni sono più amichevole e possono sopraponersi, e qualchevolte non hanno massa (Sandryanath Bose non era Italiano, ma indio). Ï bosoni hanno spin con valori interi, i fermioni hanno spin semi-intero. La maggior parte degli fermioni seguono qualche regole formulate dal inglese Dirac (l'uomo più taciturno nella storia della fisica), ma Majorana ha intuito già in 1937 che forse ci fossero particelle incapaci di generare densità di carica-corrente elettromagnetica, e ora l'ipotesi si è rivelato corretto – tale perticelle esistono. E forse anche il neutrino è una Majorana. Almeno non hanno antiparticelle – sono i suoi proprii antiparticelle. Ma forse già conosciamo le majorane - sotto forma di neutrini. È molto difficile cacciare neutrini perché non hanno carica e se hanno massa dev'essere piccolissima. Ogni momento passano miliardi di neutrini attraverso di noi senza che sentiamo niente, e tutte majorane sono altrettanto elusive perché non interagiscono con nient'affatto salvo mecchanismi basati su il loro spin e - forse - massa. E persino la famosa materia scura "rientra nella rosa dei candidati – avvantaggiata peraltro dal dettaglio non trascurabile che nessuno ha la benché minima idea di che cosa sia" (si prega notare il nodo di frase!). Qualche animi fantasiosi hanno un progietto futuristico colle majorane: "usarli come qubit nei computer quantistici. In particolare, come qubit in grado di sfruttare un fenomeno, niente affatto intuitivo, che deriva dal cosiddetto principio di sovrapposizione: assumere contemporaneamente il valore zero e uno". BAH. Se queste particelle sono così poco disposti a comunicare con noi si potrei rischiare che sia impossibile trovarla anche con una lampada e strumenti di misura elettrici. Non voglio un computer che si perde e non risponda ai miei tentativi di comunicazione - forse viaggiarei anche attraverso la Terra utilizzando l'effetto tunnel e si nasconderei da qualche parte in Cina? E se riusciremmo catturare una majorana, con ogni probabilità direbbe sì e no allo stesso tempo - proprio come i politici. Edited by Iversen on 11 November 2014 at 1:24pm 1 person has voted this message useful | |
| Iversen Super Polyglot Moderator Denmark berejst.dk Joined 6492 days ago 9078 posts - 16473 votes Speaks: Danish*, French, English, German, Italian, Spanish, Portuguese, Dutch, Swedish, Esperanto, Romanian, Catalan Studies: Afrikaans, Greek, Norwegian, Russian, Serbian, Icelandic, Latin, Irish, Lowland Scots, Indonesian, Polish, Croatian Personal Language Map | Message 3744 of 3959 13 November 2014 at 10:17am | IP Logged |
|
In May 2009 and again in November 2013 I published overviews over the results of my wordcounting extravagancies. And now I have for my own amusement and uncontrollable curiosity produced a new round of wordcounts, though this time based on micro samples of around 200 words per count (otherwise it would have taken too much time from my other activities). As usual the most relevant information is given in the form of percentages, whereas the differences in size between the dictionaries, changes in my selection criteria and hapax in the random coice of pages make the absolute numbers fairly unreliable. But if I know a third of whatever I count then that's something tangible you can use for comparisons - leaving only the caveat regarding the small sample sizes which may lead to skewed results. NB: factors 2+3 also explain the differences in total number of words estimated for one and the same dictionary.
Columns: test year and language - known words - borderline words - unknown words - total words - dictionary editions and comments 2009 Danish: 98% (58800) - … ( … ) - 2% (1200) - total: 60000 - Retskrivningsordbog 2014 Danish: 99% (51500) - … ( … ) - 2% (500) - total: 52000 - Retskrivningsordbog 2014 Norwegian: 59% (17000) - … ( … ) - 22% (6000) - total: 28000 - Paludan incl. Nynorsk Ordliste (2p NewN,, 3p. Bokmål ) 2014 Norwegian: 95% (20000) - … ( … ) - 1% (200) - total: 21000 - Pons 2009 Swedish: 70% (12100) - … ( … ) - 30% (5300) - total: 17400 - Gyldendal striped 2009 Swedish: 67% (19400) - … ( … ) - 33% (9600) - total: 29000 - Langenscheidt 2009 Swedish: 36% (30800) - … ( … ) - 64% (55100) - total: 85900 - Gyldendal red (maybe an official, but unlikely total) 2013 Swedish: 67% (11000) - 5% (1000) - 28% (4000) - total: 16000 - Gyldendal striped 2013 Swedish: 72% (25000) - 4% (1000) - 24% (8000) - total: 34000 - Langenscheidt 2013 Swedish: 48% (35000) - 9% (7000) - 43% (30000) - total: 72000 - Gyldendal red 2014 Swedish: 60% (43000) - 18% (13000) - 22% (16000) - total: 72000 - Gyldendal red 2009 Icelandic: 53% (10500) - … ( … ) - 47% (9300) - total: 19800 - Sigurdsson 2009 Icelandic: 48% (10800) - … ( … ) - 52% (11600) - total: 22400 - Iðunn 2013 Icelandic: 43% (9000) - 13% (3000) - 44% (8000) - total: 20000 - Sigurdsson 2013 Icelandic: 42% (8000) - 12% (2000) - 46% (8000) - total: 18000 - Iðunn 2014 Icelandic: 54% (12500) - 19% (4500) - 28% (6000) - total: 23000 - Iðunn 2009 German: 78% (29500) - … ( … ) - 22% (8500) - total: 38000 - Pons (Ger->sth.) 2009 German: 51% (41400) - … ( … ) - 49% (39000) - total: 80400 - Gyldendal 1999 ( (incl. expressions?)) 2012 German: 74% (46000) - 13% (8000) - 13% (8000) - total: 62000 - Gyldendal 1999 2012 German: 85% (54000) - 11% (8000) - 4% (2000) - total: 64000 - Gyldendal 1971 2013 German: 61% (37000) - 12% (7000) - 27% (17000) - total: 61000 - Gyldendal 1999 2014 German: 75% (39000) - 10% (5000) - 14% (8000) - total: 52000 - Gyldendal 1978 2014 German: 76% (41000) - 13% (7000) - 11% (6000) - total: 54000 - Gyldendal 1999 2009 Low German: 66% (4900) - … ( … ) - 34% (2500) - total: 7400 - Sass 2013 Low German: 60% (5500) - 9% (2000) - 31% (1500) - total: 9000 - Sass 2014 Low German: 69% (4000) - 11% (500) - 19% (1000) - total: 6000 - Sass 2009 Dutch: 50% (22600) - … ( … ) - 50% (22400) - total: 45000 - Gyldendal 2013 Dutch: 52% (15000) - 5% (2000) - 43% (12000) - total: 29000 - Prisma 2013 Dutch: 50% (20000) - 8% (3000) - 42% (17000) - total: 40000 - Gyldendal 2014 Dutch: 77% (31000) - 11% (4000) - 13% (5000) - total: 40000 - Gyldendal 2014 Dutch: 68% (20000) - 16% (4500) - 16% (4500) - total: 29000 - Prisma 2009 Afrikaans: 36% (16300) - … ( … ) - 64% (29300) - total: 45600 - Pharos school 2014 Afrikaans: 66% (30000) - 18% (8000) - 16% (7500) - total: 46000 - Pharos school 2006 English: 78% (35000) - … ( … ) - 22% (10000) - total: 45000 - Gyldendal 2003 (offically 45000 headwords) 2009 English: 92% (27600) - … ( … ) - 8% (2400) - total: 30000 - Oxford Advanced 2009 English: 91% (43500) - … ( … ) - 9% (4500) - total: 48000 - Gyldendal 2003 2009 English: 31% (51600) - … ( … ) - 69% (114300) - total: 165900 - Webster unabridged 2012 English: 78% (29000) - 8% (3000) - 14% (5000) - total: 37000 - Gyldendal 1971 2012 English: 66% (39000) - 10% (6000) - 24% (14000) - total: 59000 - Oxford 2013 English: 70% (23000) - 5% (2000) - 25% (8000) - total: 33000 - Oxford concise (article headwords) 2013 English: 64% (47000) - 13% (10000) - 23% (17000) - total: 74000 - Oxford concise (everything) 2013 English: 68% (36000) - 6% (3000) - 26% (14000) - total: 53000 - Gyldendal 2003 2014 English: 78% (35000) - 10% (3000) - 12% (6000) - total: 45000 - Oxford concise (one-word headwords) 2009 French: 73% (19100) - … ( … ) - 27% (7200) - total: 26300 - Micro Robert 2009 French: 51% (23100) - … ( … ) - 49% (21900) - total: 45000 - Gyldendal 1999 2010 French: 63% (28200) - … ( … ) - 37% (16800) - total: 45000 - Gyldendal 1999 2013 French: 62% (15000) - 9% (2000) - 29% (7000) - total: 24000 - Micro Robert 2013 French: 60% (21000) - 11% (4000) - 29% (10000) - total: 35000 - Gyldendal 1971 2014 French: 76% (19000) - 7% (2000) - 17% (4000) - total: 25000 - Gyldendal 1933 2014 French: 73% (26000) - 6% (2000) - 21% (7000) - total: 36000 - Gyldendal 1971 2009 Portuguese: 44% (21900) - … ( … ) - 56% (28100) - total: 50000 - Porto Editora 2009 Portuguese: 63% (12500) - … ( … ) - 37% (7500) - total: 20000 - Oxford Univ.Press (0ffic. 20000) 2009 Portuguese: 71% (19800) - … ( … ) - 29% (8000) - total: 27800 - Langenscheidt 2012 Portuguese: 67% (10000) - 11% (2000) - 22% (3000) - total: 15000 - Oxford 2012 Portuguese: 67% (20000) - 15% (5000) - 18% (5000) - total: 30000 - Langenscheidt 2014 Portuguese: 58% (15500) - 6% (1500) - 36% (10000) - total: 27000 - Langenscheidt 2014 Portuguese: 56% (28000) - 10% (5000) - 34% (9000) - total: 49000 - Porto Editora 2009 Spanish: 44% (17600) - … ( … ) - 56% (22400) - total: 40000 - Langenscheidt 2009 Spanish: 41% (19900) - … ( … ) - 59% (28100) - total: 48000 - Gyldendal 2009 Spanish: 17% (34400) - … ( … ) - 83% (165600) - total: 200000 - Bratli 2013 Spanish: 64% (21000) - 7% (2000) - 29% (10000) - total: 33000 - Langenscheidt 2013 Spanish: 42% (19000) - 8% (4000) - 50% (23000) - total: 46000 - Gyldendal 2013 Spanish: 60% (17000) - 6% (2000) - 34% (9000) - total: 28000 - Berlingske 2014 Spanish: 58% (24000) - 12% (5000) - 30% (13000) - total: 42000 - Gyldendal 2009 Catalan: 78% (13600) - … ( … ) - 22% (3900) - total: 17500 - Larousse 2013 Catalan: 70% (13000) - 9% (2000) - 21% (3000) - total: 18000 - Larousse 2013 Catalan: 55% (22000) - 8% (3000) - 37% (14000) - total: 39000 - Enciclopèdia (dE) 2014 Catalan: 71% (11000) - 11% (2000) - 18% (3000) - total: 16000 - Larousse 2014 Catalan: 66% (25000) - 12% (5000) - 22% (8000) - total: 38000 - Enciclopèdia (dE) 2009 Italian: 67% (29000) - … ( … ) - 33% (14000) - total: 43000 - new Gyldendal 2012 Italian: 63% (26000) - 14% (6000) - 23% (9000) - total: 41000 - new Gyldendal 2012 Italian: 72% (8000) - 12% (1000) - 16% (3000) - total: 12000 - D'Agostini 2013 Italian: 69% (10000) - 11% (2000) - 20% (2000) - total: 14000 - D'Agostini 2013 Italian: 63% (20000) - 6% (2000) - 31% (10000) - total: 32000 - Garzanti 2013 Italian: 60% (23000) - 9% (3000) - 31% (13000) - total: 39000 - Gyldendal 2014 Italian: 53% (23000) - 17% (7000) - 30% (13000) - total: 43000 - old Gyldendal 2014 Italian: 65% (26000) - 12% (5000) - 23% (9000) - total: 39000 - Garzanti 2007 Romanian: 16% (8000) - … ( … ) - 84% (42000) - total: 50000 - Academiei? (Offic. 50000) 2009 Romanian: 33% (13300) - … ( … ) - 67% (26700) - total: 40000 - Teora (Offic. 40000) 2009 Romanian: 43% (21400) - … ( … ) - 57% (28600) - total: 50000 - Academiei 2013 Romanian: 31% (13000) - 5% (2000) - 64% (27000) - total: 42000 - Teora 2013 Romanian: 45% (7000) - 4% (1000) - 51% (7000) - total: 15000 - Edit.Scîintifica 2013 Romanian: 15% (9000) - 2% (1000) - 83% (50000) - total: 60000 - Academiei 2014 Romanian: 62% (29000) - 9% (4000) - 29% (13000) - total: 46000 - Teora 2009 Latin: 43% (8900) - … ( … ) - 57% (11800) - total: 20700 - New College 2009 Latin: 50% (10800) - … ( … ) - 50% (10700) - total: 21500 - Langenscheidt (incl proper names?) 2013 Latin: 30% (8000) - 4% (1000) - 66% (18000) - total: 27000 - New College 2013 Latin: 34% (7000) - 9% (2000) - 57% (12000) - total: 21000 - Langenscheidt 2013 Latin: 46% (6000) - 5% (500) - 49% (6500) - total: 13000 - Gl.Gyldendal 2014 Latin: 63% (8500) - 17% (2000) - 20% (2500) - total: 13000 - Langenscheidt 2014 Latin: 56% (10000) - 13% (2500) - 31% (5500) - total: 18000 - New College 2009 Greek: 29% (7900) - … ( … ) - 71% (19200) - total: 27100 - Pataki 2009 Greek: 32% (10600) - … ( … ) - 68% (22400) - total: 33000 - new Langenscheidt 2013 Greek: 23% (8000) - 11% (4000) - 66% (23000) - total: 35000 - Pataki 2013 Greek: 18% (7000) - 5% (2000) - 77% (30000) - total: 39000 - Old Langenscheidt 2014 Greek: 51% (14000) - 13% (3000) - 35% (10000) - total: 27000 - Pataki 2013 Polish: 27% (9000) - 11% (4000) - 62% (22000) - total: 35000 - Pons 2013 Polish: 24% (3000) - 8% (1000) - 68% (10000) - total: 14000 - Oxford 2014 Polish: 44% (15000) - 8% (3000) - 48% (17000) - total: 35000 - Pons 2009 Russian: 40% (10800) - … ( … ) - 60% (16200) - total: 27000 - Langenscheidt 2009 Russian: 37% (14400) - … ( … ) - 63% (24600) - total: 39000 - Gyldendal 2013 Russian: 31% (9000) - 7% (2000) - 62% (18000) - total: 29000 - Gyldendal 2014 Russian: 50% (18000) - 10% (4000) - 40% (14000) - total: 36000 - Gyldendal 2014 Serbian: 33% (4000) - … ( … ) - 51% (8000) - total: 12000 - Сазвежћа (-wordlist) (Cyr. (done before wordlists)) 2014 Serbian: 67% (8000) - … ( … ) - 22% (4000) - total: 12000 - Сазвежћа (+wordlist) (Cyr. (done after wordlists)) 2014 Serbian: 67% (7500) - 11% (1100) - 22% (3400) - total: 12000 - Сазвежћа (alle) (Cyr. (after wordlists)) 2014 SerboCroatian: 49% (12500) - 10% (3500) - 41% (10000) - total: 26000 - Medicinska Knjiga (Latin.) 2013 Esperanto: 73% (7000) - 10% (1000) - 17% (2000) - total: 10000 - Teach yourself 2014 Esperanto: 76% (8000) - 9% (1000) - 16% (1500) - total: 10500 - Teach yourself 2013 Indonesian: 22% (2700) - 3% (500) - 75% (8800) - total: 12000 - Tuttle 2014 Indonesian: 36% (3500) - 6% (500) - 58% (6000) - total: 10000 - Tuttle Edited by Iversen on 15 November 2014 at 11:59pm 3 persons have voted this message useful |
|  | 
Printable version
|
| Copyright 2024 FX Micheloud - All rights reserved No part of this website may be copied by any means without my written authorization. |
