Re: major libgweather Locations updates
Hi, Le samedi 16 août 2008, à 13:12 +0700, Theppitak Karoonboonyanan a écrit : On Sat, Aug 16, 2008 at 11:59 AM, Theppitak Karoonboonyanan [EMAIL PROTECTED] wrote: Besides, I wonder what was the reason behind having 3 duplicated entries for VTBD (Don Muang), one for Ban Don Muang (1) (this is weird, as nobody normally calls the Don Muang district, where the Bangkok airport is located, this name any more; it sounds like very old style), another for Bangkok, and then Krung Thep. Yeah, Bangkok and Krung Thep are the same city. The former is English name, while the latter is Thai name. But what's the purpose of providing both? So, the only issue left for Thailand is whether to remove some duplicated entries. Hrm. Weird. Worth a bug report, I'd say :-) Vincent -- Les gens heureux ne sont pas pressés. ___ gnome-i18n mailing list gnome-i18n@gnome.org http://mail.gnome.org/mailman/listinfo/gnome-i18n
Re: major libgweather Locations updates
On Wed, Aug 20, 2008 at 5:40 PM, Vincent Untz [EMAIL PROTECTED] wrote: Hi, Le samedi 16 août 2008, à 13:12 +0700, Theppitak Karoonboonyanan a écrit : On Sat, Aug 16, 2008 at 11:59 AM, Theppitak Karoonboonyanan [EMAIL PROTECTED] wrote: Besides, I wonder what was the reason behind having 3 duplicated entries for VTBD (Don Muang), one for Ban Don Muang (1) (this is weird, as nobody normally calls the Don Muang district, where the Bangkok airport is located, this name any more; it sounds like very old style), another for Bangkok, and then Krung Thep. Yeah, Bangkok and Krung Thep are the same city. The former is English name, while the latter is Thai name. But what's the purpose of providing both? So, the only issue left for Thailand is whether to remove some duplicated entries. Hrm. Weird. Worth a bug report, I'd say :-) Thanks for reminding. I've already fogotten this one while translating other packages. :-/ Bug #548630 filed. Cheers, -- Theppitak Karoonboonyanan http://linux.thai.net/~thep/ ___ gnome-i18n mailing list gnome-i18n@gnome.org http://mail.gnome.org/mailman/listinfo/gnome-i18n
Re: major libgweather Locations updates
On Sat, Aug 16, 2008 at 11:59 AM, Theppitak Karoonboonyanan [EMAIL PROTECTED] wrote: Although a bit late on this, I'm examining the provided locations in Thailand. However, before filing bugs against it, I wonder what reference source you used for generating Locations.xml.in, so I can refer to similar information for missing info as stated in the comment: !-- Could not find information about the following stations, which may be in Thailand: VTBO VTCT VTPO VTPP VTSC VTSF VTSG VTSM VTUI VTUO VTUV VTUW -- (Some of these are major cities, which got removed from previous version.) Update: Having checked the stations above, I find they all have been removed from gweather for several versions, probably due to missing weather info from the server. So, just ignore them for now. Besides, I wonder what was the reason behind having 3 duplicated entries for VTBD (Don Muang), one for Ban Don Muang (1) (this is weird, as nobody normally calls the Don Muang district, where the Bangkok airport is located, this name any more; it sounds like very old style), another for Bangkok, and then Krung Thep. Yeah, Bangkok and Krung Thep are the same city. The former is English name, while the latter is Thai name. But what's the purpose of providing both? So, the only issue left for Thailand is whether to remove some duplicated entries. No new strings for now. Phew! Regards, -- Theppitak Karoonboonyanan http://linux.thai.net/~thep/ ___ gnome-i18n mailing list gnome-i18n@gnome.org http://mail.gnome.org/mailman/listinfo/gnome-i18n
Re: major libgweather Locations updates
On Tue, Aug 12, 2008 at 2:18 AM, Dan Winship [EMAIL PROTECTED] wrote: So we're not planning to revert Locations.xml.in. However, we will also try to not make any more major changes to it in 2.24. (If people find things that are horrifically broken, or that are regressions from 2.22, we'll fix them, but if they find things that are just not-as-perfect- as-they-could-be, we'll wait.) Although a bit late on this, I'm examining the provided locations in Thailand. However, before filing bugs against it, I wonder what reference source you used for generating Locations.xml.in, so I can refer to similar information for missing info as stated in the comment: !-- Could not find information about the following stations, which may be in Thailand: VTBO VTCT VTPO VTPP VTSC VTSF VTSG VTSM VTUI VTUO VTUV VTUW -- (Some of these are major cities, which got removed from previous version.) Wikipedia [1] has some info, but the coordinates seem to be a little different from the ones in the existing data. [1] http://en.wikipedia.org/wiki/List_of_airports_by_ICAO_code:_V#VT_-_Thailand Besides, I wonder what was the reason behind having 3 duplicated entries for VTBD (Don Muang), one for Ban Don Muang (1) (this is weird, as nobody normally calls the Don Muang district, where the Bangkok airport is located, this name any more; it sounds like very old style), another for Bangkok, and then Krung Thep. Yeah, Bangkok and Krung Thep are the same city. The former is English name, while the latter is Thai name. But what's the purpose of providing both? Regards, -- Theppitak Karoonboonyanan http://linux.thai.net/~thep/ ___ gnome-i18n mailing list gnome-i18n@gnome.org http://mail.gnome.org/mailman/listinfo/gnome-i18n
Re: major libgweather Locations updates
Sorry about not getting back to this sooner. I talked with Vincent about this (with him wearing his libgweather maintainer hat, not his release team hat), and I think we agree that the possible scenarios are something like this, ranked from best to worst: 1. Keep new Locations.xml.in, all translation teams fully translate it 2. Keep new Locations.xml.in, translation teams translate at least the new locations in countries where their language is widely spoken 3. Revert to old Locations.xml.in, use existing translations 4. Keep the new Locations.xml.in, no new translations Right now we're somewhere around 4, but it's not a whole lot of effort to get to 2, which I think is all we need to target for 2.24; no one using the Arabic localization is going to notice if there are small towns in Denmark left untransliterated. We also agreed that it might make sense to remove libgweather (at least the po-locations part) from the translation statistics, for exactly that reason; it's not like with ordinary UI strings, where any user in any language is equally likely to encounter any string. So we're not planning to revert Locations.xml.in. However, we will also try to not make any more major changes to it in 2.24. (If people find things that are horrifically broken, or that are regressions from 2.22, we'll fix them, but if they find things that are just not-as-perfect- as-they-could-be, we'll wait.) -- Dan ___ gnome-i18n mailing list gnome-i18n@gnome.org http://mail.gnome.org/mailman/listinfo/gnome-i18n
Re: major libgweather Locations updates
Le mardi 05 août 2008 à 22:59 -0400, Dan Winship a écrit : Claude Paroz wrote: Le lundi 04 août 2008 à 07:37 -0400, Dan Winship a écrit : Dan, do you have criteria to include new cities? For example, for Liechtenstein, a very small country of 160 km2 and 35000 inhabitants, there is 11 cities. What's the use case? The vast Liechtensteinian GNOME hacker contingent? Important enterprise desktop sales among Lichtenstienian megacorporations? Until now, I was only aware of the French-domination plot. This one seems a more secret one :-P Hm... so basically this was the script being dumb. One of the criteria I was using to decide major cities was cities that are the capitals of top-level administrative divisions (eg, states provinces, etc), because our dataset has that information for most countries, but it doesn't have population information. This heuristic worked really well in some countries, but then in other places it looks like it's just filling Locations.xml up with crap... So I took out that rule and regenerated Locations.xml.in, and now 10 of those 11 cities in Liechtenstein are gone, along with 619 other cities around the world. (A couple dozen of those cities actually *are* major cities that we want to eventually put back, but we can fix those later.) Thanks for fixing this. Claude ___ gnome-i18n mailing list gnome-i18n@gnome.org http://mail.gnome.org/mailman/listinfo/gnome-i18n
Re: major libgweather Locations updates
Djihed Afifi wrote: First, are you absolutely sure this is stable now? I won't touch this until it is. No, that's why I suggested people should wait a week or so before making major effort at translating it. Because as people report problems, I'm going to try to make improvements. I am not sure what methodology was used to prune or consider cities. For the most part, it is the same as before; we list cities where there are weather stations. The only difference is that now: - We also try to include major cities that don't have their own weather stations - We try to have each location be a city name, rather than an airport name or something else - If there is a location in a small town, and there's a medium-sized city nearby, we try to name the location after the medium-sized city instead of the small town. (This depends on us having good data about the relative sizes of cities, so it works better in some countries than others.) Fundamentally, things are still driven by where there are weather stations reporting. I had a quick look at Arab cities in general and Algerian cities in particular. So as a first pass, you sould just be comparing to GNOME 2.22, rather than to the theoretical perfect list of Algerian cities. You'll find that the list hasn't changed all that much (and hopefully where it has changed, it's better now. Eg, Algiers has been added, Tamanrasset/Aguenna (a combined city and airport name) has now become Tamanrasset (just the city name), etc. Algeria also has a very odd population density. 80% of people live in 15% of the land. But more than half the cities in the file are Saharan cities. Yeah, so this is because that's where the weather stations are. From googling/wikipedia-ing, it looks like a lot of them are associated with oil fields. If they really are basically useless, we can remove them from the list. (Vincent and I were having this debate on IRC the other day, about a handful of weather stations in very very tiny towns (eg, ~100 people) in France. On the one hand, it seems better to keep them--maybe the computers at the oil field run GNOME? :) But on the other hand, more locations means more work for translators, so I can understand the desire to keep the list small...) The city choice was not made on population either. For countries where we have population data, all cities with population greater than 100,000 are included. However, we don't have population data for very many countries. We can fix this manually by adding cities to libgweather/data/major-cities.txt. In brief, some Algerian choosing cities, and looking under Algeria will be wondering how exactly were the cities chosen. I suspect the same will be true for many people around the world. Yes, but that was true in GNOME 2.22 too. This is a first step toward improving it. As for your question of what to do with suggested improvements, please don't make changes directly to Locations.xml.in at this time; put the fixes in a bug report instead. -- Dan ___ gnome-i18n mailing list gnome-i18n@gnome.org http://mail.gnome.org/mailman/listinfo/gnome-i18n
Re: major libgweather Locations updates
What I fear is that this is just replacing a fundamentally flawed list with another fundamentally flawed list (even if less so). Are you sure you don't want to take the changes offline, make sure that you make it as good as possible, take a year if needed, then upload the almost perfect one? On the one hand, I understand that getting it released and receiving bug reports is important, on the other, it just seems to me that ~1800 strings is a bit too much a price in terms of manpower to have that. That's at least 10 hours * 50 active teams = 500 hours. That's a lot. With ~ 1 month left (effectively three weeks after the first one week period) to fully translate it. I wish you guys would consult translators first for such big changes. Djihed On Wed, Aug 6, 2008 at 4:57 PM, Dan Winship [EMAIL PROTECTED] wrote: Djihed Afifi wrote: First, are you absolutely sure this is stable now? I won't touch this until it is. No, that's why I suggested people should wait a week or so before making major effort at translating it. Because as people report problems, I'm going to try to make improvements. I am not sure what methodology was used to prune or consider cities. For the most part, it is the same as before; we list cities where there are weather stations. The only difference is that now: - We also try to include major cities that don't have their own weather stations - We try to have each location be a city name, rather than an airport name or something else - If there is a location in a small town, and there's a medium-sized city nearby, we try to name the location after the medium-sized city instead of the small town. (This depends on us having good data about the relative sizes of cities, so it works better in some countries than others.) Fundamentally, things are still driven by where there are weather stations reporting. I had a quick look at Arab cities in general and Algerian cities in particular. So as a first pass, you sould just be comparing to GNOME 2.22, rather than to the theoretical perfect list of Algerian cities. You'll find that the list hasn't changed all that much (and hopefully where it has changed, it's better now. Eg, Algiers has been added, Tamanrasset/Aguenna (a combined city and airport name) has now become Tamanrasset (just the city name), etc. Algeria also has a very odd population density. 80% of people live in 15% of the land. But more than half the cities in the file are Saharan cities. Yeah, so this is because that's where the weather stations are. From googling/wikipedia-ing, it looks like a lot of them are associated with oil fields. If they really are basically useless, we can remove them from the list. (Vincent and I were having this debate on IRC the other day, about a handful of weather stations in very very tiny towns (eg, ~100 people) in France. On the one hand, it seems better to keep them--maybe the computers at the oil field run GNOME? :) But on the other hand, more locations means more work for translators, so I can understand the desire to keep the list small...) The city choice was not made on population either. For countries where we have population data, all cities with population greater than 100,000 are included. However, we don't have population data for very many countries. We can fix this manually by adding cities to libgweather/data/major-cities.txt. In brief, some Algerian choosing cities, and looking under Algeria will be wondering how exactly were the cities chosen. I suspect the same will be true for many people around the world. Yes, but that was true in GNOME 2.22 too. This is a first step toward improving it. As for your question of what to do with suggested improvements, please don't make changes directly to Locations.xml.in at this time; put the fixes in a bug report instead. -- Dan ___ gnome-i18n mailing list gnome-i18n@gnome.org http://mail.gnome.org/mailman/listinfo/gnome-i18n
Re: major libgweather Locations updates
This immense change is something that I personally do not approve of. Translating cities, huts, hamlets and jerk water somewhere is no particular fun. And there is no particular use of this. I as a coordinator of the Bulgarian Gnome project am putting translation of the places in gweather in the trash. The developers of gweather can scrape atlases and maps for whatever damn, uninhabited place on Earth (or Moon actually, why the smeg is Mare Tranquillitatis missing from gweather?) or stranded military base. Life is way too short for this shit. The least they could have done is to have provided us with IPA transcriptions. I cannot think of any motif for their deed but the direst passion to be blogged about in linuxhaters. Come on guys, you could have contacted the team there directly, no need for such convoluted ways. This is not the incredible machine, you are not Rube Goldberg. al_shopov ___ gnome-i18n mailing list gnome-i18n@gnome.org http://mail.gnome.org/mailman/listinfo/gnome-i18n
Re: major libgweather Locations updates
On Tue, 2008-08-05 at 21:42 +0300, Александър Шопов wrote: This immense change is something that I personally do not approve of. Translating cities, huts, hamlets and jerk water somewhere is no particular fun. And there is no particular use of this. I as a coordinator of the Bulgarian Gnome project am putting translation of the places in gweather in the trash. The developers of gweather can scrape atlases and maps for whatever damn, uninhabited place on Earth (or Moon actually, why the smeg is Mare Tranquillitatis missing from gweather?) or stranded military base. Life is way too short for this shit. The least they could have done is to have provided us with IPA transcriptions. I cannot think of any motif for their deed but the direst passion to be blogged about in linuxhaters. Come on guys, you could have contacted the team there directly, no need for such convoluted ways. This is not the incredible machine, you are not Rube Goldberg. al_shopov Please keep your rants to yourself, or just send to linuxhaters directly. This kind of language is *not* appreciated. Neither by gweather developers, nor by other translators, or any other member of the project, or really, any sane person. -- behdad http://behdad.org/ Those who would give up Essential Liberty to purchase a little Temporary Safety, deserve neither Liberty nor Safety. -- Benjamin Franklin, 1759 ___ gnome-i18n mailing list gnome-i18n@gnome.org http://mail.gnome.org/mailman/listinfo/gnome-i18n
Re: major libgweather Locations updates
Am Dienstag, den 05.08.2008, 21:42 +0300 schrieb Александър Шопов: I cannot think of any motif for their deed but the direst passion to be blogged about in linuxhaters. Though I'm also not in favor of this, I'd prefer to keep this discussion rational. The reason for this commit does have been explained: http://mail.gnome.org/archives/gnome-i18n/2008-August/msg00046.html andre -- mailto:[EMAIL PROTECTED] | failed http://www.iomc.de/ | http://blogs.gnome.org/aklapper ___ gnome-i18n mailing list gnome-i18n@gnome.org http://mail.gnome.org/mailman/listinfo/gnome-i18n
Re: major libgweather Locations updates
2008/8/5 Behdad Esfahbod [EMAIL PROTECTED] On Tue, 2008-08-05 at 21:42 +0300, Александър Шопов wrote: This immense change is something that I personally do not approve of. Translating cities, huts, hamlets and jerk water somewhere is no particular fun. And there is no particular use of this. I as a coordinator of the Bulgarian Gnome project am putting translation of the places in gweather in the trash. The developers of gweather can scrape atlases and maps for whatever damn, uninhabited place on Earth (or Moon actually, why the smeg is Mare Tranquillitatis missing from gweather?) or stranded military base. Life is way too short for this shit. The least they could have done is to have provided us with IPA transcriptions. I cannot think of any motif for their deed but the direst passion to be blogged about in linuxhaters. Come on guys, you could have contacted the team there directly, no need for such convoluted ways. This is not the incredible machine, you are not Rube Goldberg. al_shopov Please keep your rants to yourself, or just send to linuxhaters directly. This kind of language is *not* appreciated. Neither by gweather developers, nor by other translators, or any other member of the project, or really, any sane person. Agreed. -- behdad http://behdad.org/ Those who would give up Essential Liberty to purchase a little Temporary Safety, deserve neither Liberty nor Safety. -- Benjamin Franklin, 1759 ___ gnome-i18n mailing list gnome-i18n@gnome.org http://mail.gnome.org/mailman/listinfo/gnome-i18n ___ gnome-i18n mailing list gnome-i18n@gnome.org http://mail.gnome.org/mailman/listinfo/gnome-i18n
Re: major libgweather Locations updates
Le lundi 04 août 2008 à 07:37 -0400, Dan Winship a écrit : Andre Klapper wrote: Am Sonntag, den 03.08.2008, 21:24 -0400 schrieb Dan Winship: I've just committed a huge update to libgweather's Locations.xml.in. From an i18n perspective, the big changes are that a lot of strings representing airport names, etc, went away and were replaced with actual city names. Also, many city names in some countries were replaced with better-localized/better-transliterated versions. Uhm. This has decreased all translations by 5-6% which is A LOT: Yup. But most of the new untranslated strings are either: (a) the names of minor cities that aren't going to have different names in different languages anyway (though I realize these still require transliterations in non-latin-alphabet languages) Dan, do you have criteria to include new cities? For example, for Liechtenstein, a very small country of 160 km2 and 35000 inhabitants, there is 11 cities. What's the use case? Claude ___ gnome-i18n mailing list gnome-i18n@gnome.org http://mail.gnome.org/mailman/listinfo/gnome-i18n
Re: major libgweather Locations updates
On Aug 5, 2008, at 21:46 , Behdad Esfahbod wrote: On Tue, 2008-08-05 at 21:42 +0300, Александър Шопов wrote: This immense change is something that I personally do not approve of. Translating cities, huts, hamlets and jerk water somewhere is no particular fun. And there is no particular use of this. I as a coordinator of the Bulgarian Gnome project am putting translation of the places in gweather in the trash. The developers of gweather can scrape atlases and maps for whatever damn, uninhabited place on Earth (or Moon actually, why the smeg is Mare Tranquillitatis missing from gweather?) or stranded military base. Life is way too short for this shit. The least they could have done is to have provided us with IPA transcriptions. I cannot think of any motif for their deed but the direst passion to be blogged about in linuxhaters. Come on guys, you could have contacted the team there directly, no need for such convoluted ways. This is not the incredible machine, you are not Rube Goldberg. al_shopov Please keep your rants to yourself, or just send to linuxhaters directly. This kind of language is *not* appreciated. Neither by gweather developers, nor by other translators, or any other member of the project, or really, any sane person Well, I share his desperation with this change, and while his use of words is telling things in too harsh or inappropriate ways for many, he is kind of right. The real question is why we were translating all kind of obscure military base names in the first place, and not the approximate place names which would make much more sense to anyone living around the area. Not trying to blame anyone, it would have been huge task to do this right from the beginning, only thing is that now the huge task of fixing things is multiplied by languages wishing to keep things fully translated 100%. This might be the most important case of 'get your original translations right to start with' we have ever seen. Personally, I got tired with gcompris doing pretty much same thing regularly, you get 20-40% of fuzzy or untranslated strings when the change is just change in wording or adding a comma or dot somewhere. Gcompris is developed by French people and I understand the reasons why it was not perfect to start with, but with gcompris I got too tired follow all the changes. I promise to do it 'soon' but... Our team, finnish, will follow do the tedious task of redoing all of gweather before release, I just want to remind everyone: try to write good English whenever you write some terms to translate. Any change you need to do to original strings mean useless extra work for hundreds of translators, even if it's just 'oh they added a comma to the phrase' *hile* ___ gnome-i18n mailing list gnome-i18n@gnome.org http://mail.gnome.org/mailman/listinfo/gnome-i18n
Re: major libgweather Locations updates
Le mardi 05 août 2008, à 22:56 +0300, Ilkka Tuohela a écrit : Our team, finnish, will follow do the tedious task of redoing all of gweather before release, I just want to remind everyone: try to write good English whenever you write some terms to translate. Any change you need to do to original strings mean useless extra work for hundreds of translators, even if it's just 'oh they added a comma to the phrase' (note that by not cc'ing Dan, you won't get replies from him) While you're right, the libgweather case is different and not related to bad english. It's due to the fact that there's no good and consistent source of data for locations in the world. Dan's work is the attempt to fix this. I can understand the frustration that it can cause. On the other hand, we don't have to take into account this translation in statistics -- I think we ignored it in the past, eg. Vincent -- Les gens heureux ne sont pas pressés. ___ gnome-i18n mailing list gnome-i18n@gnome.org http://mail.gnome.org/mailman/listinfo/gnome-i18n
Re: major libgweather Locations updates
Claude Paroz wrote: Le lundi 04 août 2008 à 07:37 -0400, Dan Winship a écrit : Dan, do you have criteria to include new cities? For example, for Liechtenstein, a very small country of 160 km2 and 35000 inhabitants, there is 11 cities. What's the use case? The vast Liechtensteinian GNOME hacker contingent? Important enterprise desktop sales among Lichtenstienian megacorporations? Hm... so basically this was the script being dumb. One of the criteria I was using to decide major cities was cities that are the capitals of top-level administrative divisions (eg, states provinces, etc), because our dataset has that information for most countries, but it doesn't have population information. This heuristic worked really well in some countries, but then in other places it looks like it's just filling Locations.xml up with crap... So I took out that rule and regenerated Locations.xml.in, and now 10 of those 11 cities in Liechtenstein are gone, along with 619 other cities around the world. (A couple dozen of those cities actually *are* major cities that we want to eventually put back, but we can fix those later.) -- Dan ___ gnome-i18n mailing list gnome-i18n@gnome.org http://mail.gnome.org/mailman/listinfo/gnome-i18n
Re: major libgweather Locations updates
On Mon, Aug 4, 2008 at 2:24 AM, Dan Winship [EMAIL PROTECTED] wrote: I've just committed a huge update to libgweather's Locations.xml.in. From an i18n perspective, the big changes are that a lot of strings representing airport names, etc, went away and were replaced with actual city names. Also, many city names in some countries were replaced with better-localized/better-transliterated versions. This is unfortunately a huge amount of churn in an already-huge list of translatable strings. Two suggestions: 1) Search the po file comments for This is the capital of, which will let you find national capitals, which are presumably more likely than average to need special translations. Also, search for is the traditional English name to find cities that are called something different in English than they are in the local language, which usually also points to the need for a translation. 2) Other than that, wait a week or so (or more) before translating anything, and don't clean out the now-unused translations right away, because we'll be requesting some help from gnome-love which may result in some lame cities being removed and other ones being renamed, etc. Also, once GNOME 2.23.6 is out there, we're going to bump the intltool requirement to the latest version, which will let us add msgctxts to disambiguate duplicate names (eg the US state of Georgia vs the former Soviet Republic of Georgia). For Greece I have noticed the following, a. The names in the original strings now have accents, such as Alexandroúpolis, which helps non-native speakers. This is cool, but also means that more messages require attention for subtle changes. I think it's a move to the right direction. b. There are multiple entries for airports, having them associated with several nearby cities. For example, for LGKV, there are entries for four cities, Chrysoúpolis, Dráma, Kavála, Xánthi. http://maps.google.com/maps?f=qhl=engeocode=q=%CE%BA%CE%B1%CE%B2%CE%AC%CE%BB%CE%B1,+%CE%B5%CE%BB%CE%BB%CE%AC%CE%B4%CE%B1sll=34.741612,-95.625sspn=68.827616,113.203125ie=UTF8ll=41.109365,24.535217spn=0.512179,0.884399t=pz=10 How much is the radius you picked when considering nearby cities? Is it more than 50Km? c. In some cases, the transliterated version does not have an accent. For example, Chrysoúpolis Airport and Chrysoupoli Airport It looks as if the airport name of the original entry gets the transliteration, while the nearby cities do not get transliteration of the airport name. Simos ___ gnome-i18n mailing list gnome-i18n@gnome.org http://mail.gnome.org/mailman/listinfo/gnome-i18n
Re: major libgweather Locations updates
2008/8/4 Andre Klapper [EMAIL PROTECTED] Am Sonntag, den 03.08.2008, 21:24 -0400 schrieb Dan Winship: I've just committed a huge update to libgweather's Locations.xml.in. From an i18n perspective, the big changes are that a lot of strings representing airport names, etc, went away and were replaced with actual city names. Also, many city names in some countries were replaced with better-localized/better-transliterated versions. Uhm. This has decreased all translations by 5-6% which is A LOT: http://l10n.gnome.org/releases/gnome-2-24 I'd prefer to revert this change and get it in early for 2.25 so teams have enough time to translate - now it's only 7 weeks left and a lot of other modules to translate for most teams, too. But I'm not a translator so I'd like to see opinions of other teams. I'm from the Danish translation team. Normally I'd prefer to just push through and get big changes done right away, BUT this one is not only a big change but also one that will require lots of research, so I agree with Andre that it would be better to commit this patch early in the next release cycle. Regards Kenneth Nielsen ___ gnome-i18n mailing list gnome-i18n@gnome.org http://mail.gnome.org/mailman/listinfo/gnome-i18n
Re: major libgweather Locations updates
Andre Klapper wrote: Am Sonntag, den 03.08.2008, 21:24 -0400 schrieb Dan Winship: I've just committed a huge update to libgweather's Locations.xml.in. From an i18n perspective, the big changes are that a lot of strings representing airport names, etc, went away and were replaced with actual city names. Also, many city names in some countries were replaced with better-localized/better-transliterated versions. Uhm. This has decreased all translations by 5-6% which is A LOT: Yup. But most of the new untranslated strings are either: (a) the names of minor cities that aren't going to have different names in different languages anyway (though I realize these still require transliterations in non-latin-alphabet languages) (b) the names of cities in countries where GNOME doesn't currently have many (or *any*) users (because in the places where there are lots of GNOME users, people had already gone through and replaced airport names with city names and added entries for major cities years ago, so there aren't as many changes there now). So while the translation stats are now much worse on paper, they're worse in ways that our current users probably won't notice. The reason we wanted to get this in for 2.24 is that the new autocompleting GWeatherLocationEntry works better if everything is based on city names, because it has the user typing their location in rather than picking it from a giant list, and they're not going to type the name of a random airport or geographical feature nearby, so with the old Locations.xml.in, it would be harder for people to find the right location. -- Dan ___ gnome-i18n mailing list gnome-i18n@gnome.org http://mail.gnome.org/mailman/listinfo/gnome-i18n
Re: major libgweather Locations updates
On Mon, Aug 4, 2008 at 12:52 PM, Dan Winship [EMAIL PROTECTED] wrote: Simos Xenitellis wrote: For Greece I have noticed the following, a. The names in the original strings now have accents, such as Alexandroúpolis, which helps non-native speakers. This is cool, but also means that more messages require attention for subtle changes. I think it's a move to the right direction. We could optionally strip the diacritics out for 2.24 to reduce the number of new strings, and then bring them back post-2.24? My preference would be to keep the transliterated version with diacritics. Sorry for not being clear. Thanks for the work, Simos b. There are multiple entries for airports, having them associated with several nearby cities. For example, for LGKV, there are entries for four cities, Chrysoúpolis, Dráma, Kavála, Xánthi. http://maps.google.com/maps?f=qhl=engeocode=q=%CE%BA%CE%B1%CE%B2%CE%AC%CE%BB%CE%B1,+%CE%B5%CE%BB%CE%BB%CE%AC%CE%B4%CE%B1sll=34.741612,-95.625sspn=68.827616,113.203125ie=UTF8ll=41.109365,24.535217spn=0.512179,0.884399t=pz=10 How much is the radius you picked when considering nearby cities? Is it more than 50Km? It's exactly 50km. That may need adjusting. (If we lower it, than some of those cities will just go away entirely; it always uses the closest weather station for a city, so the fact that those cities use the LGKV code means there isn't any other weather station closer to them than that.) (In this case, Chrysoúpolis is listed because that's where the LGKV weather station actually is, and Dráma, Kavála and Xánthi are listed because they're marked as being the capitals of states/provinces/ whatever-you-call-them-in-Greece in our dataset, which is one of the things we use to guess that a city is a major city that should be listed.) c. In some cases, the transliterated version does not have an accent. For example, Chrysoúpolis Airport and Chrysoupoli Airport It looks as if the airport name of the original entry gets the transliteration, while the nearby cities do not get transliteration of the airport name. Hm... yeah, that's a bug (Chrysoupoli Airport is the original weather station name from the METAR source file, Chrysoúpolis Airport is obviously the more-fixed-up version). But in this case it doesn't matter, because that string never gets displayed in the UI. (You'll note that neither version shows up in the .po file.) The METAR source strings only get used in cases where there is more than one weather station in a city (which doesn't seem to be the case for anywhere in Greece). -- Dan ___ gnome-i18n mailing list gnome-i18n@gnome.org http://mail.gnome.org/mailman/listinfo/gnome-i18n
Re: major libgweather Locations updates
Am Sonntag, den 03.08.2008, 21:24 -0400 schrieb Dan Winship: I've just committed a huge update to libgweather's Locations.xml.in. From an i18n perspective, the big changes are that a lot of strings representing airport names, etc, went away and were replaced with actual city names. Also, many city names in some countries were replaced with better-localized/better-transliterated versions. #: ../data/Locations.xml.in.h:40 msgid Abilene Regional Airport #: ../data/Locations.xml.in.h:50 msgid Abumusa Island, Abumusa Airport #: ../data/Locations.xml.in.h:769 msgid Babelthuap Island, Babelthuap /Koror Airport #: ../data/Locations.xml.in.h:2301 msgid Cincinnati Municipal Airport, Lunken Field Is there a concept behind the different versions of listing an airport (commas, no commas) that I don't get? ;-) andre -- mailto:[EMAIL PROTECTED] | failed http://www.iomc.de/ | http://blogs.gnome.org/aklapper ___ gnome-i18n mailing list gnome-i18n@gnome.org http://mail.gnome.org/mailman/listinfo/gnome-i18n
Re: major libgweather Locations updates
Andre Klapper wrote: #: ../data/Locations.xml.in.h:40 msgid Abilene Regional Airport #: ../data/Locations.xml.in.h:50 msgid Abumusa Island, Abumusa Airport #: ../data/Locations.xml.in.h:769 msgid Babelthuap Island, Babelthuap /Koror Airport #: ../data/Locations.xml.in.h:2301 msgid Cincinnati Municipal Airport, Lunken Field Is there a concept behind the different versions of listing an airport (commas, no commas) that I don't get? ;-) The closest there is to a rule is that if there's a comma, then everything on the same side of the comma as the word Airport is part of the airport name, and anything not on the same side probably isn't. Here's the situation: our primary source of weather station information is just the most godawful database in the world, with absolutely no consistency in naming conventions, ridiculous misspellings and typos, out-of-date names for airports (and even for cities in some places), etc. You can see my efforts to improve the situation here: http://svn.gnome.org/viewvc/libgweather/trunk/data/station-fixups.pl?view=markup In the case of the specific examples you gave, the names of the airports themselves are Abilene Regional Airport, Abumusa Airport, Babelthuap/Koror Airport (the weird spacing around the / is a bug), and Cincinnati Municipal Airport. The middle two are preceded by the name of the island where they are, which maybe I should be stripping out because the information probably isn't that useful. The last one is suffixed with the name of a specific area within the airport, which I should *definitely* be stripping out... (Note that these strings are only used to differentiate different weather stations within a city. Eg, someone living in Cincinnati can pick either Cincinnati Municipal Airport or Cincinnati / Northern Kentucky International Airport depending on which is closer to them, and get better weather reports that way.) -- Dan ___ gnome-i18n mailing list gnome-i18n@gnome.org http://mail.gnome.org/mailman/listinfo/gnome-i18n
Re: major libgweather Locations updates
2008/8/4 Dan Winship [EMAIL PROTECTED] Andre Klapper wrote: Am Sonntag, den 03.08.2008, 21:24 -0400 schrieb Dan Winship: I've just committed a huge update to libgweather's Locations.xml.in. From an i18n perspective, the big changes are that a lot of strings representing airport names, etc, went away and were replaced with actual city names. Also, many city names in some countries were replaced with better-localized/better-transliterated versions. Uhm. This has decreased all translations by 5-6% which is A LOT: Yup. But most of the new untranslated strings are either: (a) the names of minor cities that aren't going to have different names in different languages anyway (though I realize these still require transliterations in non-latin-alphabet languages) (b) the names of cities in countries where GNOME doesn't currently have many (or *any*) users (because in the places where there are lots of GNOME users, people had already gone through and replaced airport names with city names and added entries for major cities years ago, so there aren't as many changes there now). So while the translation stats are now much worse on paper, they're worse in ways that our current users probably won't notice. Yeah but the thing is though, that the translation statistics is all we have to gauge our performance on. It may sound childish I don't know, but this page http://l10n.gnome.org/languages/da/gnome-2-24 is what we use both as motivation and as goal during the big translation update for gnome 2.24, and seeing those UI stats go to 99.5% is just nowhere as exiting or fulfilling as seeing them go to 100%. On another note, seeing the size of this change makes me wonder. It must have taken an immence development effort to complete this update, streching over quite some time, would it really have been so difficult to warn us back when you began and then to commit every once in a while so we could have had more time with some of the strings? The reason we wanted to get this in for 2.24 is that the new autocompleting GWeatherLocationEntry works better if everything is based on city names, because it has the user typing their location in rather than picking it from a giant list, and they're not going to type the name of a random airport or geographical feature nearby, so with the old Locations.xml.in, it would be harder for people to find the right location. That does sound like an improvement that _might_ warrant this change at this time, I cant really judge that and so I won't object to it staying in but I wont applaude it either. I can only regret that it was not done differently. Kenneth Nielsen ___ gnome-i18n mailing list gnome-i18n@gnome.org http://mail.gnome.org/mailman/listinfo/gnome-i18n
