Thanks for the info. I hope you will get an answer soon - there is so much to do! I look forward to continue.

Still no sign of a response - I guess they are busy suing New Line Cinema at the moment:)

---

sí tere hyelle ar nullave cenilve

I must admit that the same idea crossed my mind when I read the news :)

The real question is, what will I do if they do _not_ bother to answer (I guess we will just carry on and take it as a yes:))

---

sí tere hyelle ar nullave cenilve

I'm already making sure to use EQ and Gn roots rather than LQ and N/S ones. It just happens that in PE12:31 the EQ root for "iron" is given as ANGA, which is similar to ANGÂ in Etym. I already knew not to categorize EQ with LQ! I assumed the EQ root would be the same as used in Goldogrin, as the Lexicon didn't show a root there.

As for _ang_ linking to _anga_, I actually was trying to link it to EQ _anga_, since "iron" has the same word in EQ and LQ - I didn't think there would already be a page for LQ _anga_, so I just linked it to "anga".

That said, what should we do about words that clearly survive all the way from Goldogrin to Noldorin to Sindarin, like _thanc_ or _ang_? Though they're from different time periods, should their pages be somehow linked to indicate that the words are related?

When linking Noldorin words to "Middle" Quenya (Qe) ones, instead of checking if the wiki contains that particular Qe word aleady I rather make sure that the link contains the reference (Qe), i.e. Qe [[aika (Qe)|aika]]. If Qe aika is in the wiki, the two entries are linked together automatically; if it is not and the link remains red, then we at least see that it is missing and has to be entered sometime.
As to the words that survive all the way from Goldogrin through ON, N (even Ilk, Dor, etc) to Sindarin I would suggest to add a reference to the entries, e.g. G ang Cf. N ang, S ang. The links would then look like this: N [[ang (N)|ang]], S [[ang (S)|ang]].

Yes, if a words survives in subsequent conceptional stages they should be cross-referenced - like the cited early Qenya anga (EQ), where I put in

Just for the archives: all entry names/links in EW should have the language marker in brackets, e.g. anga (EQ), anga (Qe), anga (LQ). If you put in a link to. e.g., anga (LQ), but you do not want the (LQ) part showing on the page, you can achieve this by putting [anga (LQ)|] in the source. The 'pipe' character | tells the wiki engine to cut the text in brackets from the diplayed text but leaves it for the link.

The reason for all this magic is that each every page is stored in a separate record in the underlying database. As the title of the page (= the name of the entry) identifies the record, naturally all page titles have to be different.

---

sí tere hyelle ar nullave cenilve

In addition to using the pipe, I would kindly ask all wiki co-workers to please put the language abbreviations (Qe, LQ, ON, Dor etc.) before the Eldarin words when cross-referencing.

Okay, thanks - I was wondering how it was possible to link to a page with (LQ) in the title but not have it appear in the link!

As for the roots, will a similar thing be done with them, i.e. ANGA (EQ) and ANGÂ (Qe)?

Something similar, yes - I am open to suggestions there:) It is clear that roots should have their higher metacategories, too: roots from the EQ-Goldogrin era, Etym roots, later roots which are usually classified as Common Eldarin in the sources, etc.

---

sí tere hyelle ar nullave cenilve

Hi,
In order to keep the Wiki uniform and to avoid the duplication of categories, the best way is to copy/paste the existing entries from the language semantic categories. But this means a lot of clicking around in the Wiki. I have therefore made a list of the sources, the general language categories and the Buck'categories in a word document. This allows a faster copying/pasting than to navigate around the Wiki. If you like I can mail it to you.

Erm, you shouldn't really use those anymore... unless we want a total confusion over the wiki. We've come to a decision to use Buck's low-level fields on the entry-level and Buck's high-level categories on the language-level. Maybe if you look at a few entries in the Early Qenya or Qenya section it will become clearer, because those entries have already been made according to these "rules".

But if you absolutely "must" have a look at the old semantic categories, they are still visible if you go through Categories - Semantic categories (but they will gradually disappear as we are moving all entries from them to the new categories).

Complicated, eh?

---

sí tere hyelle ar nullave cenilve

Much more so taking into consideration that it was _us_ (e.g. you Atwe and me) who told Cerebrum a couple of days ago to still use the old semantic categories... now _I'm_ confused :)

A good decision Tyrhael!
Well, I can add new glyphs to the font, no problem with that. I would appreciate if you compiled a comprehensive list of all special characters required incl. minuscules/majuscules. In the meantime you could go on with editing entries which don't require any of these, just to keep wiki going :)
Waiting to hearing from you!

Okay; I'm going to try and edit/add more Wiki stuff when I get home later today. Whenever I come across a special character needed, I'll add it in a list. I'm going to mainly be working with cognates in a sort of web — for example, add all the stuff with _ain-_, one of which is _Angainos_, which contains _ang_, then I'll move on to everything containing _ang_, etc. etc.

Well, yes. I mean there has been some publicity around it, like Ryszard (Galadhorn) kindly put a banner and link up in his Elendili forum, also I shamelessly smuggled a link into the relevant entries over on Wikipedia. I have also emailed some people I _thought_ might be interested in. They were, sort of, just never popped up as actual contributors.
But if you have good tips, please do not hold yourself back...

---

sí tere hyelle ar nullave cenilve

for not answering earlier, I forget sometimes we have a discussion page over here :)
I agree with you, we cannot avoid adding new fields. As you say, Buck selected Indo-European synonyms and we should use this as a basis trying to stick as close as possible to it, but expanding it where necessary.
Progress is very slow indeed, especially on my side. Whenever I am not sure of a field I start reading through all the synonyms and that is very time consuming. Also I am having doubts when splitting up entries (like 'helca' the other day). Maybe you could have a look at it to see if it is the way you want it. Then, either I forget the cognates or I am not sure if it is correct what I entered (have not entered many).
I hoped you would make a presentation about the Wiki on Tatya. I think this is a topic as interesting as any other and maybe we can get some help - we have nearly 3 days to convince them! Well, I am an optimist!

I agree.

---

sí tere hyelle ar nullave cenilve

That would be of big help probably, so thank you in advance!

There are some fields in the list that start like "Notes on words blahblah...". I assume these have meanings in the context of the book, but look odd in the wiki, so I usually amend them and take the Notes on... part out, if you agree.

Also, if you make an "other" field in any of the metacategories, pls do not call it just "Other":), make it "Time, other" or similar.

---

sí tere hyelle ar nullave cenilve

Notes on ... give more detailed information on specific entries as you can see from the one I copied for you on 13.192 Note on other words for a collective body. I do not mind if you take those sub-categories out.

"Other" field : I am not aware of such a field. Can you tell which one, so I can correct. I take normally the metacategories' name or the field name: Emotion - other / House - other.

You misunderstand me - I did not mean to delete the subcategories, just to delete the "Notes on" part from the name of the subcategory, as it does not make sense when you are looking at an entry.

Other: e.g. under Quantity and Number there was the last subcat. 13.53 Miscellaneous, I have corrected it to 13.53 Quantity and Number, Miscellaneous.

But these are only small things.

Shine on!

---

sí tere hyelle ar nullave cenilve

:) For 13.53 you will have to complain to Mr. Buck. It is indeed just
13.53 Miscellaneous !
But you are right to complete it. If you do not mind, we could adapt it to 13.53 Quantity and Number - other, for the sake of conformity.

Notes on.. I have taken that out.

I agree with you, we will have entries that are difficult to place. So generally, I think we should put such entries into "garbage" categories (xxx - other), rather than leave them aside.

Specifically, "and" fits etymologically "beside, next", but it is a conjunction and there are no such fields in any of the categories. So I would put it in an "other" field.

Other, but other what? What is 'and' in a semantic sense?

---

sí tere hyelle ar nullave cenilve

The meaning of "and" is "thereon", "next" and as a conjunction adds one element (words, phrases) to an other one. Another meaning is more specifically "plus" like in two and three = 5. I would, for the time being, put it into "Quantity, Number - other". If this is not completely right, then it is certainly not totally wrong either :)

Hmm, now that you mention it... thanks for the brainstorming:)

---

sí tere hyelle ar nullave cenilve

You probably know that there is already a topic/thread here on Aglardh precisely to collect such possible connections, so I suggest to put them in there. For the time being let's keep the wiki as factual as we can get, and keep original research separate, just like Wikipedia does (with which I do not want to say that we cannot speculate on the meanings of words here and there but that is still within the realm of Eldarin linguistics).
Then if we can gather a considerable corpus of possibly connected words there will be no reason not to make a separate project or even a subproject of Aelfwina Boc to deal with it.

---

sí tere hyelle ar nullave cenilve

You are right, I forgot that thread. For the moment, I shall note them and if I gather a certain number of them put them there periodically.

I think it would be interesting! But I'd rather create a new link for it on the wiki main page, the Sandbox should remain the playground for trying out the wiki features etc. Atwe, what's your opinion?

(we have yet to decide on the structure of the language-related semantic categories);

Did I interpret something wrong? I thought we had decided to use the 23 metacategories for the language related ones and have started to enter them. Maybe I was to hasty?

No, you're right, sorry, it was an unfortunate choice of words. What I wanted to say was that AFAIK we haven't decided whether to use the category numbers or the titles yet. But I think the way you've started entering the language-related categories is perfectly fine, carry on, I will follow in your footsteps :)

I'm just having a bit of trouble adding cognates and roots, as many of the entries say "see such and such" in the GL but I only have PE13 and not either Lexicon. I mean, I can take _afasaith_ and find AWA "wealth" from BoLT 1, and look for _saith_ 'hunger' and any related words ...

As for similar words that aren't cognates: should I just link them to the same root that they're derived from; i.e. I shouldn't link to Ausir, ausie, etc. in the _afasaith_ page, but only on AWA2?

And I'm going to have to look into the coding of already-made pages a bit more to get an idea of how it's done.

I'm just having a bit of trouble adding cognates and roots, as many of the entries say "see such and such" in the GL but I only have PE13 and not either Lexicon. I mean, I can take _afasaith_ and find AWA "wealth" from BoLT 1, and look for _saith_ 'hunger' and any related words ...


Erm... I was rather thinking of cognates in other languages such as early Quenya from QL maybe -- the problem is, I don't have PE13 and have no idea what it contains exactly or how it is presented. Therefore I would suggest you only add cognates where any given. It is perfectly okay if an entry is only a "stub" for the time being. To be more specific let me tell you how I create ON, N, PQ, CE entries from scratch: I am currently working based on the Etym, I just tackle the roots one by one and create the relevant entries. If a word (e.g. N Aglond) appears first under the root AK-, then I create the entry according to what information I find under AK-. Later on, when I reach root LOD-, I extend/modify/make notes/etc. the entry "Aglond" accordingly. The same will happen if I find more (or even different) information on "Aglond" in one of the HoME volumes, VTs, etc. What I want to say is that an entry is practically never complete when it is first created, so don't put any pressure on yourself by trying to gather all information at once. This goes for the roots as well: if there is none mentioned in PE13 and you can't be reasonably sure from which root the given word derives, no problem, leave it out, you'll probably find it in another source sometime. Or others may and add this info to the entry.


As for similar words that aren't cognates: should I just link them to the same root that they're derived from; i.e. I shouldn't link to Ausir, ausie, etc. in the _afasaith_ page, but only on AWA2?


I don't link together words deriving from the same root as a rule only if appropriate or specifically mentioned by Tolkien. Taking your example, I may refer to "Ausir" in the "au(s)saith" (and not "afasaith") page if relevant, but I wouldn't refer to all words derived from AWA2 (cf. Noldorin entry "agr"). Since you create the root category AWA2 in all relevant entries, if you then check AWA2 in the roots category, you'll find all pertinent entries listed there (try checking an existing root).


Don't loose faith, it only sounds complicated, in fact it isn't (or is it?) :)

I uploaded some more words; Gnomish forms of Ainu and Aini (Ainil, Ain, Ainos, Ainuil, Ainog) and some words on cheese and curdling — verbs, nouns, and adjectives. I think I'm getting the hang of it, but can anyone check to see that I'm not starting any bad habits that would continue throughout all the other entries I'm going to create? For example, there's a word _cír_ that's a noun "cheese" in one place and an adjective "sour" on another; should I try and make two pages, or place them as two separate entries on one like I have it now? Sorry for the hassle, but I plan on uploading more and more and want to stop bad habits before they _become_ habits. ;)

I would give them separate entries, distinguishing them with 1 and 2 in superindex, like e.g. the ar- entries in the Qenya section.

I noticed that you still use the "old" semantic categories - please read our discussion here in this topic about the introduction of Buck's semantic categories and how we use them.

---

sí tere hyelle ar nullave cenilve

- generally the best way to understand the structure of the pages is to look at the source of the page by hitting Edit
- the category pages are basically normal wiki pages, with the addition that the wiki software automatically lists all pages linked to the category on the category page

The MediaWiki Handbook is a good reference to formatting, linking etc. syntax and rules

---

sí tere hyelle ar nullave cenilve

14.31 - 14.35 list adverbs of time, I think we could add 14.352 after, 14.353 before or maybe grouping them into one number = 14.352 time prepositions.

For the spatial relations, I see the following possibilities:
(after 12.33 Top)
12.332 on
12.333 over

(after 12.36 Side)
12.362 in front of
12.363 behind

or (after 12.38 Center)
12.382 in front of
12.383 behind

Buck has sub-numbers for similar meanings like End, Point, Edge, so we could adopt the same system and place additional fields where it seems most logic.

What about the language related semantic category? Do we take the metacategories in a short form (EQ 1. Physical World, EQ 2. Mankind, EQ 3. Animals, EQ 4. Body, EQ 5. Food, EQ 6. Clothing, EQ 7. Dwelling, EQ 8. Agriculture, EQ 9. Physical Acts, etc.?

All right, just let us not fall into the trap of creating a new subcategory for every single word we encounter:)

Language-level: yes, I think the 23 metacategories will be enough, and also consistent with the general tree.

---

sí tere hyelle ar nullave cenilve

I think we should decide something because the missing fields will be a major problem. If it is quite obvious that most fields can receive various words related to the field (mountain = mountain peak, mountain cleft etc) the same is not possible for verbs and related nouns and vice versa, because the field specifies (vb.) or (noun). What do you suggest? The easiest could be to create at the end of each chapter an additional number that contains various chapter related words, e.g. under 14. Time = 14.81 Sundry time related words (or some similar title). That would make only on category more per chapter.

How should we shorten the title of no. 19?

After having entered a few categories, I suddenly have a doubt. Do we use Buck's categories as he does or do we just let us guide by them and use them more or less as we did with the old categories. That means: Buck analyses the word of a given field not various words pertaining to it - 19.22 A people, nation renders the words used in different languages for people or nation. So we would have the word lië for Quenya in this field and we would have to open a new field for nations' names. Or do we use this field to enter the various nations: Falyanel etc.?

I can see your problem there. The way I see it at the moment is that Buck's categories were made to fit a basic common vocabulary and its different mutations in the various Indo-European languages. So as it is now it seems that not all Eldarin words will fit into one of the existing subcategories. I think that the specific subcategories should be kept for the words that specifically fit into them - i.e. into the field "A people, nation" all the Eldarin words denoting people, nation, folk etc. should go, but not the specific names of the various peoples like Eldar, Falathrim etc.
Of course that will leave us with a host of words that do not fit into this basic wordlist. With them we have two choices:
- if you feel that the word is part of the basic Middle-Earth vocabulary then add a subcategory for it - like I did e.g. for "swan", for which Buck didn't have any fields but which I feel features prominently in Tolkien's legendarium, so I added it under "Animals" - but I would not add, of course, a category for _ammale_
- for all the others maybe we should create "garbage-categories" under the main categories, like you have proposed, e.g. "The physical world - other" etc.

- BTW I think for the nations and peoples - as their names are proper nouns - we can create a subcat under Names - "Names of peoples".

---

sí tere hyelle ar nullave cenilve

I know the Quenya disambiguation is currently going on, but what's happening with Goldogrin? There isn't a category for it, and that's what the majority of the words I have catalogued are so far, other than early Qenya; I just got my copy of PE13 in the mail as well, with a large amount of Goldogrin and early Noldorin.

Tyrhael, I will create a new category box on the front page of the wiki in the evening. Would you be willing to start creating entries?

I'll play in the Sandbox a little to make sure I have the process down, then start uploading from PE13; I'll leave the EQ and Gold. from BoLT 1 & 2 for later. So I can upload, then they can be added in a category for Gnomish after it's created?

Absolutely! I am glad that you are willing to contribute. Everybody has only so much time on his/her hands, so all help is fantastic.

If it weren't for the bots who register in the wiki and upload Viagra-links to the first page they find I would de-protect the Main Page so that the users can add languages.

Thanks
---

sí tere hyelle ar nullave cenilve

I added two articles: _afasaith_ and _avosaith_, can someone check that they're formatted and written how they be should so I can continue the same way with other entries? Is there something I have to do after writing the article, i.e. finding the category pages and putting it in manually there? For example, I added it to AWA2 "fortune, wealth" from the QL/GL which doesn't yet exist; should I go to the roots page and add AWA2?

I'm afraid we will face this problem again and again... but hey, that's the beauty of it :) Just musing: what about using either 19.21 People (Populace) or 9.23 Tribe, Clan, Family ? Does it make any sense?

Since we keep the whole hierarchy tree, we should keep the numbers - therefore either

15.71 Touch (vb.)
or
Semantic: 15.71 Touch (vb.)

I feel we could go for the simpler form (15.71 Touch (vb.) ) because anyone looking for semantic categories will see how it is organized and know once and for all that the numbered categories are semantic ones.

I agree with going for the simpler form as suggested above. If we keep the sequence of the categories in each entry (lexical-semantic-source-root) there will be no confusion about which is which, and the entries look more tidier as well.

OK, I have started working on the Q entries again, using the category:X.YY Blah format in the general semantic categories.
Unfortunately the old semantic categories will only disappear when we have finished rehauling the whole Q section.

I have also added a link and explanation to EditingSuggestions; I suggest that you all use simple copy/paste to copy the category.names first to preserve consistency.

The hierarchy I set up is Categories - semantic categories - general semantic categories/language-specific semantic categories(to be added later) - the 22 metacategories á la Buck + 1 Names - the semantic fields.

---

sí tere hyelle ar nullave cenilve

I also added a Names metacategory, with subfields, please check and edit as desired: 23._Names

---

sí tere hyelle ar nullave cenilve

Watching this mighty collection of categories unfurl over in the Sandbox I really start thinking that we should simply keep them as they are, maybe add a few subcategories here and there if needed (and the one main category for the names). Granted, there weill be subcategories that will never be used because we simply lack the vocabulary, but then they will just be that - left empty.űűThe thing we have to decide on is that beside putting the words in the general semantic fields do we also want to have language-level semantic categorization - maybe just use the main categories on language-level? (otherwise we'd end up with categories containing a single Qenya etc. word).

---

sí tere hyelle ar nullave cenilve

Obviously, I do not think either that we should use that many categories, but that we could break each chapter into maybe 3 or 4 subcategories, where we deem it necessary. While typing I noticed that the subcategories are numbered very logically.
Let's take e.g. in 4. Parts of the body, the most easily recognizable as a group:
4.20 - 4.29 contains words about the head down to the throat.
4.30 - 4.393 words of articulated members
4.51 - 4.67 bodily functions
4.71 - 4.84 body conditions

So chapter 4 could be broken up like its title says into "Parts of the body", "Functions" and "Conditions". All the chapters are grouped that way (well I must admit sometimes I fail to understand why a word figures in one number group rather than in another).

I think the main interest of all the 22 pages that I am entering is in fact to follow a professional uniform categorization. You will notice that we had entered e.g. words into "geography" while Buck has them in "motion" (street, paths etc.).

Buck's very detailed categorization should just help us decide which word belongs to which category.
The advantage is that we can easily break up a category if we find it too crowded following one of his groupings.

I would find it interesting to have both a general semantic category and a semantic language related category.

I, on the other hand think that in the general semantic tree we should retain as much details as possible (perhaps simply adopting the Buck-hierarchy as is). The reason for this is when you research the languages in a comparative way you are probably looking how they express a single notion/idea/thing; e.g. by looking at how Qenya and Noldorin express "dog" you may form an idea how it might look in e.g. Telerin even if it is non-attested.

On the other hand, in the language-level semantic hierarchy no great level of detail is required, because you are probably interested in, say, all the names of the colours in Noldorin, and not just one specific colour, in a category.

---

sí tere hyelle ar nullave cenilve

I agree, my post was written with language related categories in mind where it does not make sense to go into too much detail. If we have two categorizations, the general semantic one should indeed be more detailed otherwise it gets again too crowded with all the different languages. Adopting the whole Buck's hierarchy for the general semantic cat. means we do not have to rack our brains with grouping his sub-categories and for the language related one applies what I explained before: main categories and only a few sub-categories especially where we can make logic groups.
Luckily, I have another day off tomorrow. So I shall be able to enter the last ones. Glad to finish (big book and the most horribly small print :-(