Überarbeitung der Wortliste

Moderator: Moderatoren

Benutzeravatar
Lúthien Meliel
Beiträge: 79
Registriert: Do Jun 26 2008 21:58
Kontaktdaten:

Beitrag von Lúthien Meliel » Mo Okt 17 2011 8:05

Ah thanks! I had indeed missed some of that.

I'd be in favor of adding everything which could be potentially helpful.
People can always switch them off if they don't want to use them for whatever reason.

If this is felt like "treading on uncertain ground" we could make it so that the goodies are disabled by default, leaving it to the individual user to enable them after having clicked through various disclaimer screens ;)

Bild

But seriously: I think it's a good idea to leave the choice to use features like are discussed here to the discretion of the user. I know that I'd be delighted to have them available. When I originally asked Didier whether we could include the PE17 words he didn't want to do that for reasons I never really understood - but it was something like "that PE 17 is a dictionary on its own". I think that our user population is very well able to make those decisions for themselves; and if it made clear *what is what*, it isn't going against the linguistic integrity of the material.

Technically spoken it's trivial to create a switch option between "linguistically prudent" and "include bells & whistles".

Benutzeravatar
Eryniel Elmíris
Beiträge: 1520
Registriert: Di Jun 24 2008 19:51
Wohnort: Ribobel / Rímdor

Beitrag von Eryniel Elmíris » Mo Okt 17 2011 9:03

LOL

But seriously...
Technically spoken it's trivial to create a switch option between "linguistically prudent" and "include bells & whistles".
I like that approach.

Benutzeravatar
Roman
Admin
Beiträge: 2018
Registriert: Di Mär 13 2007 13:27
Kontaktdaten:

Beitrag von Roman » Mo Okt 17 2011 20:08

I think that our user population is very well able to make those decisions for themselves; and if it made clear *what is what*, it isn't going against the linguistic integrity of the material.
Yeah, well.. In theory, that's true. But I've experienced that people are much influenced by a dictionary anyway. You can see the hypercorrected **revia- 'fly' (instead of rhenia- or sindarized renia-) and **panna- 'fill' (instead of pannad-) all over the internet including scholarly articles, despite clear indications in Hiswelóke that these have been altered in some way. I myself used them, too.
Although maybe the problem with regularizations actually is that the comments indicating alteration are separated from the words themselves. Perhaps we need a new symbol prefixed to the glosses, like !renia- or something...

Benutzeravatar
Lúthien Meliel
Beiträge: 79
Registriert: Do Jun 26 2008 21:58
Kontaktdaten:

Beitrag von Lúthien Meliel » Mi Okt 19 2011 22:18

yes, I understand that.
Next to using such a symbol, maybe we could use something like colouring the word red to make it very obvious what the status of the word is?

Benutzeravatar
Lúthien Meliel
Beiträge: 79
Registriert: Do Jun 26 2008 21:58
Kontaktdaten:

Beitrag von Lúthien Meliel » Do Okt 20 2011 12:02

ok, in order to move ahead I should like to settle on the sort of data that can be expected.

1) It's easy to "misuse" the "reconstructed" column in order to indicate whether or not an entry is regularised.
For now, 0 means "not reconstructed / attested"; and 1 means "reconstructed". 2 could then mean "regularised".

2) For the potential future past tenses, it would be most handy to create an additional column past in the Entry table , similar to the plural and alt columns (this would allow for any number of past tense forms to be added, and the number in the past column would serve as enumeration for those forms).

3) Then there's the possible translation alternatives. I'm now discussing this with Eryniel on Skype, and I hope that I am understanding it correctly like this:
Imagine I'm searching for the english word glory (using the old Hisweloke dataset now)
This gives me the following results:

aglar S., N. ['ɑglɑr] n. (en) glory, brilliance, splendour ◇ Ety/348, S/427, LotR/II:I, LotR/VI:IV, RGEO/73, VT/47:13
claur N. [kl'ɑur] n. (en) splendour, glory ◇ Ety/362


Would, in this case, the "hints" be brilliance, splendour and splendour ?
If so, there is no need to add something to the datamodel, it is already possible.

4) Lastly, there's Eryniel's remark:
Eryniel Elmíris hat geschrieben:(...)
- And last but not least Roman wished to include the non-regularized forms where applicable, but if we would do that, then the Noldorin past-tense should also be included.
if we could settle on these points, I can finalise the dataset and start on the application itself. (past tenses etc. or whatever is mentioned above can than be added to the data later)
[/b]

Benutzeravatar
Roman
Admin
Beiträge: 2018
Registriert: Di Mär 13 2007 13:27
Kontaktdaten:

Beitrag von Roman » Do Okt 20 2011 23:19

1) It's easy to "misuse" the "reconstructed" column in order to indicate whether or not an entry is regularised.
For now, 0 means "not reconstructed / attested"; and 1 means "reconstructed". 2 could then mean "regularised".
Yes, I think we should do that.
2) For the potential future past tenses, it would be most handy to create an additional column past in the Entry table , similar to the plural and alt columns (this would allow for any number of past tense forms to be added, and the number in the past column would serve as enumeration for those forms).
Since verbs have no plural and nouns have no past, one could actually use the plural column for that, unless it would create problems elsewhere.
Would, in this case, the "hints" be brilliance, splendour and splendour ?
I don't know what 'hints' are in this context, but if you type in 'glory', you should get all the entries in full where 'glory' appears in the translations.
The thing is that you have to take the meaning of a Sindarin word from the overlap of the English translations. For example, 'appear' can be used in the sense 'seem', or in the sense 'come into sight' (as in 'Hamlet appeared on the stage'). The Sindarin word thia- has only the former meaning as seen by the gloss 'seem, appear'. If you look for 'appear' and just get thia-, you may be misled into using it in the latter sense. So the English-Sindarin wordlist is deprecated, if you will.
if we could settle on these points, I can finalise the dataset and start on the application itself. (past tenses etc. or whatever is mentioned above can than be added to the data later)
Right, so we need at least a 'verbatim gloss' column and a 'Noldorin past' column; perhaps also a 'Sindarin past' column as mentioned above. I'll fill the data in later.

I should mention btw, that I'm in the process of moving to another city to a new job; and my brains are a bit fried now between delving into a new physical subject and life-organizing shenanigans. I hope that the sandstorm will settle down in a couple of weeks, though.
I couldn't bring any Tolkien material with me, but I should have all what I need in digital form. This was partly why I swooped down on the wordlists in summer - to finish them before moving house.

P.S.
potential future past tenses
Fascinating. 8-)

Benutzeravatar
Lúthien Meliel
Beiträge: 79
Registriert: Do Jun 26 2008 21:58
Kontaktdaten:

Beitrag von Lúthien Meliel » Fr Okt 21 2011 11:15

Roman hat geschrieben: 2) ...
Since verbs have no plural and nouns have no past, one could actually use the plural column for that, unless it would create problems elsewhere.
It is certainly technically possible, but I doubt if it's worth the effort of fetching the distinction between verb and noun elsewhere (in the metadata table) in order to determine the meaning of that column.
Incidentally, this is a good example of the central dilemma of database design: the choice between making the model as small, flexible and appropriate as possible, and considerations like performance speed and readability.
Roman hat geschrieben: I don't know what 'hints' are in this context (...)
Eryniel and I were talking about 'hints' the other day. I thought that she had used that in a previous post, but I can't find it now. What we mean is "translation alternatives" like in
Eryniel hat geschrieben: - the other suggestion was to display not only the sindarin word if you search for a translation in german, but also all possible translation alternatives. For example: you enter Ruhm" as search word and would get something like: aglar - Glanz, Ruhm (sources)...
Roman hat geschrieben:but if you type in 'glory', you should get all the entries in full where 'glory' appears in the translations.
The thing is that you have to take the meaning of a Sindarin word from the overlap of the English translations. For example, 'appear' can be used in the sense 'seem', or in the sense 'come into sight' (as in 'Hamlet appeared on the stage'). The Sindarin word thia- has only the former meaning as seen by the gloss 'seem, appear'. If you look for 'appear' and just get thia-, you may be misled into using it in the latter sense. So the English-Sindarin wordlist is deprecated, if you will.
OK, that makes a lot of sense indeed. I'm thinking about what to do about it.
In the given example it's like this:

to appear ->
thia- -> to appear
thia- -> to seem

The English meaning to appear is ambivalent. Too bad there isn't a Sindarin word for the 'coming into sight' meaning of to appear .. I'm trying to think of other cases where this might be the case.
But for the time being, imagine there would be a Sindarin word that meant coming into sight - say, autha-, searching for to appear could be made to yield these results:

to appear ->
thia- -> to appear (thia-) in the meaning of 'to seem'
thia- -> to appear (autha-) in the meaning of 'to come into sight'
thia- -> to seem (thia-)

It won't be too hard to make it so, because the links are already there. It's just a matter of following it one step further.

But in the existing example, it wouldn't work because there is no Sindarin word available (yet) which has the meaning of 'to come into sight'. Still, someone might be tempted to use thia- in that sense, because of their experience with English, and not realising that meanings may cover different ground in Sindarin.
The point is that in the cases that there is no Sindarin word for the different meanings of a given English (or German) word, this information is not contained within the dataset as it is now, and it would have to be added.

Am I correct to assume that it comes down to this question: do we want to explicitly warn the user for cases like this, where there is a chance that they would use a Sindarin word in the wrong meaning of a given English/German word which also has a Sindarin equivalent with the correct meaning?
It could be added to the Commentary of the translation (the combination of a Sindarin and a German/English word); or maybe it's better to add it to the Sindarin entry itself maybe?

Roman hat geschrieben:
potential future past tenses
Fascinating. 8-)
It is, isn't it :)
Douglas Adams, in 'the Restaurant at the End of the Universe', hat geschrieben: (...)
The major problem (of time travel) is simply one of grammar, and the main work to consult in this matter is Dr. Dan Streetmentioner's Time Traveler's Handbook of 1001 Tense Formations. It will tell you, for instance, how to describe something that was about to happen to you in the past before you avoided it by time-jumping forward two days in order to avoid it. The event will be descibed differently according to whether you are talking about it from the standpoint of your own natural time, from a time in the further future, or a time in the further past and is futher complicated by the possibility of conducting conversations while you are actually traveling from one time to another with the intention of becoming your own mother or father.

Most readers get as far as the Future Semiconditionally Modified Subinverted Plagal Past Subjunctive Intentional before giving up; and in fact in later editions of the book all pages beyond this point have been left blank to save on printing costs.

The Hitchhiker's Guide to the Galaxy skips lightly over this tangle of academic abstraction, pausing only to note that the term Future Perfect has been abandoned since it was discovered not to be.
Good luck with the move!

Benutzeravatar
Lúthien Meliel
Beiträge: 79
Registriert: Do Jun 26 2008 21:58
Kontaktdaten:

Beitrag von Lúthien Meliel » Fr Okt 21 2011 12:31

Roman hat geschrieben: I don't know what 'hints' are in this context, but if you type in 'glory', you should get all the entries in full where 'glory' appears in the translations.
but that it already did in the previous incarnation:

Bild

Benutzeravatar
Roman
Admin
Beiträge: 2018
Registriert: Di Mär 13 2007 13:27
Kontaktdaten:

Beitrag von Roman » Fr Okt 21 2011 19:31

do we want to explicitly warn the user for cases like this, where there is a chance that they would use a Sindarin word in the wrong meaning of a given English/German word which also has a Sindarin equivalent with the correct meaning?
I don't think we have to that. Firstly, it's not always easy to predict where confusion might arise. And secondly, just showing the whole entries as in the picture above is more than enough disambiguation, is it not?
but that it already did in the previous incarnation:
Perfect! :-)
Also, I would like to retain the same kind of search routine on the website which can be helpful in various situations. I guess this means that sindarin.de will host the English and French dictionary as well - that's cool.

Benutzeravatar
Roman
Admin
Beiträge: 2018
Registriert: Di Mär 13 2007 13:27
Kontaktdaten:

Beitrag von Roman » Fr Nov 04 2011 12:14

Is it just me or is there an elephant in the room? I have installed it on my firefox and even though it uses a Belgian flag instead of a German one, it's much like the application we're building...

Ailinel
Beiträge: 1715
Registriert: Fr Mär 16 2007 11:53

Beitrag von Ailinel » Fr Nov 04 2011 13:36

One wonders who that "kind developer" is that contacted D. W. "early in October"...

Didier
Beiträge: 3
Registriert: Fr Nov 04 2011 14:10

Beitrag von Didier » Fr Nov 04 2011 14:40

Hello Roman

While entering new words and references from PE/18 and PE/19 in my dictionary, I cross-checked against your own helpful list (version dated 02/10/2011), so I shall have here the courtesy to post the two small differences I noticed on the way.

- Missing reference : aglar occurs in PE/18:87 (with usual meaning)
- taw vs. daw, do "thither (dorthin)" : you seem to consider them separately in your list, I rather wonder whether the latter (along with its shortened form) is not simply the mutated form of the former ? (This would perhaps even imply that this word usually occurs after verbs or nouns, hence the lenition.)

Cheers,

Didier.
P.S. I'll have the German flag fix in LBreath, sorry for that *blushes*.
P.P.S. Hic sunt dracones...

Ailinel
Beiträge: 1715
Registriert: Fr Mär 16 2007 11:53

Beitrag von Ailinel » Fr Nov 04 2011 16:13

Hic sunt dracones, quibus facultas legendi est...

Benutzeravatar
Eryniel Elmíris
Beiträge: 1520
Registriert: Di Jun 24 2008 19:51
Wohnort: Ribobel / Rímdor

Beitrag von Eryniel Elmíris » Sa Nov 05 2011 0:18

I cant comment on the application as I can't code, but it does look rather familiar...
And sorry Ailinel, I never did Latin ;)

Benutzeravatar
Roman
Admin
Beiträge: 2018
Registriert: Di Mär 13 2007 13:27
Kontaktdaten:

Beitrag von Roman » Sa Nov 05 2011 11:08

- taw vs. daw, do "thither (dorthin)" : you seem to consider them separately in your list, I rather wonder whether the latter (along with its shortened form) is not simply the mutated form of the former ? (This would perhaps even imply that this word usually occurs after verbs or nouns, hence the lenition.)
It is doubtlessly the lenited form, but I didn't know how to account for that when I started out with the very simple table. Using the search mask, the len. tag appears beside daw, do, but something like that is also in order for the html list.

Nonne est Hisveloce draco? ;-)

Antworten