Hello PKP Forum!
First of all, I would like to congratulate the people of PKP for the development and continued support of this invaluably useful set of software and all people who have contributed to PKP for their part in making this software great. I came across the PKP software collection by chance and I was immediately attracted by the clean, professional and streamlined design and its wealth of features.
I would also like to thank James MacGregor personally for creating an account for me in PKP Forum and thus solving the problem I had registering.
Cretan Studies Association, the non-profit, cultural and research institution in which I'm participating, has scheduled two conferences for this summer, one for the popular oral culture of Crete and Cyprus and one for the Greek poets Angelos Sikelianos and Odysseas Elytis, and OCS seemed to be an excellent and very powerful solution for the tasks of collecting abstracts and presentations for the conference and making them available in a functional and appealing form.
In fact, for the purposes of these two specific conferences, which are mostly built around invited presentations, OCS carries many features which we will probably not be using. Alternatively, we were considering of making use of the institution's main, Joomla based, webpage, by implementing a document management solution like Docman and/or a specially design registration-submission form.
However, I thought that it is better to plan ahead and I just didn't like not having the option available, if needed. Open Conference Systems is of course available, but it was important for us to have the platform localized, and unfortunately Greek locale was not available. A specific circumstance gave me the time opportunity of translating the necessary OCS files myself and I took it.
So here is the greek translation of the necessary OCS locale files: (for safety, always check the newest post in this thread for the latest, most up to date translation
Here is an updated translation of the en_US locale files for OCS...
You must know that this update assumes you have implemented the changes in this report/patch http://pkp.sfu.ca/bugzilla/show_bug.cgi?id=6397
It also includes some translation optimization, to render meaning more concisely
All future translation updates will be built on the basis of those changes...so If you use this translation version, you MUST implement the above changes.A few words about the translation1. What does it include?
I have translated all .xml files inside OCS/locale
, all files inside OCS/lib/pkp/locale except
"countries.xml" and "currencies.xml" (already translated) & the various locale files inside the OCS/plugins
folderEDIT - FEB 4th 2011
I have also included the Greek locale files for TinyMCE, downloaded from the editor's website...
I have also translated the hover text for the font size plugin. There was no locale file for that, so I had to edit the "fontController.js" and convert it from ANSI to UTF-8 without BOM
Most of the files within the OCS/lib/pkp/locale
directory were already translated into Greek, but I went ahead and translated them from scratch because:
a. I wanted to have a more uniform translation across OCS
b. I found some sparse errors with the existing translation
c. Translating the files directly instead of cross-checking between the existing greek and the english locale actually proved less time-consuming
d. The existing translation must have been based on an earlier OCS/OJS release and there were message-keys missing from most files
e. The grid.xml file was not included in the set of files
I was actually bored of the task of dealing with "countries.xml" and "currencies.xml", even though the former has additional strings compared to the en_US version.
I should also point out that the translation was based on the en_US
locale files (with a little help from the french and italian locales for clarification and confirmation purposes), meaning that the greek locale files are now updated and support the most recent version of OCS to date.2. What was the approach used?
Translating is never easy, even if the vocabulary itself does not present difficulties. The difficulties are rather morphological and semantic and, more precisely, they arise from the combination of these two elements. There are, also, some specific difficulties and concerns when translating from English into Greek, which are even more accentuated when dealing with software translations, where terms can be isolated and mixed randomly (from the translator's point of view) and different terms from different parts of one or more language files can be brought together in the actual working version of the software in a way that the translator cannot know before-hand, thus, lacking an immediate perception of the specific context. But for these difficulties to "translate" into translation errors, some specific syntactical and grammatical differences, hinted above, must also come into play.
So here are a couple of points on issues which might have brought about the more obvious and more difficult to prevent translation (where guessing the context becomes a crucial factor) and which have determined grammatical choices for the translation, which in turn can have stylistic or semantic consequences perhaps not fitting to everyone's taste and judgement:
1) English is quite flexible, meaning that a single term in its specific form can be used to refer to many differing terms. For example, "completed" could refer to a submission, to many submissions, to registration(s), to text field(s) etc. In Greek, the same word would differentiate grammatically depending on the gender, the singular or the plural form of the word to which it refers.
I was generally able to deal with these problems (correctly I believe) drawing on the immediate context. However, the immediate context (preceding/anteceding text, or the coding of the message-keys themselves) does not always help and there are also occasions where following those leads would actually be misleading. So there are bound
to be some instances of mistaken translation, and if it is proved there are none, then this would be solely the product of luck
With the above description, I believe I have also sufficiently hinted to the fact thatI have not tested the translation on a working OCS site (edit - February 2011: The translation has now been widely tested, but I could still have missed certain typographical errors and improvements could still be possible)
. I believe, exactly because of the above difficulties, that this is an absolutely necessary step in finalizing the translation
, in order to prevent errors (especially because these might most probably appear in the seemingly simplest of cases) or confusing messages.
But because I have taken much time dealing with the major task of the translation (and because I adopted a "bottoms up" approach to it, leaving my eyes aching a bit) and because I will not be using OCS extensively in the most immediate future and in its full width of functions, I will probably not be able to single out these potential translation problems. So, I will be needing the help and input of greek-speaking users of OCS
2) English, because of its flexibility, is advantageous when it comes to producing short messages conveying a specific meaning. In Greek, it almost invariably takes more (and I might also add longer) words in order to specify the meaning and reference of that same message. This can become a serious problem in cases where the design and layout of a software program is/might be based on implicit presuppositions about the area taken up by text. I have tasted this bitter experience with phpbb, phpbb styles, Joomla, Joomla extensions etc, where it often happens that when one switches to the greek translation, things suddenly go out of bounds, overlap, take up new lines or just make everything look weird by taking up more space etc.
As I said, I have not yet tested this translation on a working OCS site (edit - February 2011: The translation has now been widely tested, but I could still have missed typographical errors and improvements could still be possible)
, but with this concern in mind I have made some grammatical and syntactical choices which on occasions could make the translation appear rigid, maybe not immediately clear and/or scholarly (please do not read this as an ambition on my part) and/or idiosyncratic:2. a) Idiosyncrasies
- Instead of using the imperative mood and active verbal forms, I have used nouns, signifying function, e.g. "προσθήκη" (addition) instead of "προσθέστε" (add).
- I have used many participles, e.g. "ανατεθειμένη" ("assigned"), instead of "έχει ανατεθεί"
- By choosing nouns instead of verbs, I have managed to avoid the use of articles in these short messages, but I have also managed to produce sequences of nouns connected by genitive case.
The combination of the above do make up a more rigid translation for the sake of brevity and, at times, maybe at the expense of clarity.
- Based on the hierarchy of roles in OCS, I have translated a couple words different from or contrary to the usual translation of these terms in Greek. "Manager" is translated as "Διευθυντής" (and "management" - where it refers directly to the manager's actions - as "διεύθυνση") and "Director" as "Επιμελητής".
- I have translated "archive" as "ομάδα αρχείων" in order to distinguish it from "αρχείο" used for "file"
I have mumbled a lot, I know...I just wanted to make some things clear
and inform others adequately with the intent of making the best possible finalized translation.
I don't expect many typographical errors to be found, since I have checked each and every line carefully after completing the translation, but if you find any error, of any kind, no matter how great or small, or if you have a better idea about the translation of some messages, or about the translation of terms across the locale files, please reply to this thread indicating the directory location, the name of the file, the specific line and its content (including the message key).
I have found that opening the files in Notepad++ and searching within and across the contents of the files is a good way of locating errors and homogenizing the translation.