RTÉ-fada screwing up the EPG.

Solair

Why does RTE insist on sticking a fada on the É everywhere?

I understand that it might be part of their branding, but normally, diacritical marks i.e. accent marks (fadas) are only used to change the sound of a word when it is being pronounced. Nobody uses them in abbreviations, unless you are reading them phonetically.

I have a few issues with it where RTÉ shows up as RTA or shows up as RT_ !

Also, when RTÉ is displayed in monospace text, e.g. teletext or many EPGs it distorts the capital E by squashing it, which looks aweful.

I don't know why they have to be so ridiculously pedantic about the whole thing it just looks crap!

Mike 1972

Meanwhile over on analogue theyve had the wrong names on RTE2 and TG4 for well over a decade now :rolleyes:

Minstrel27

What analogue system is that?

Mike 1972

If you have an analogue tuner with a auto channel naming feature (works off Teletext signals) RTE2 is still labelled "NET 2" and TG4 is "TNAG"

Its only been a decade (and some) like..........:rolleyes:

Red Alert

The thing I don't see the need for is the on-screen RTE logo. It may have been of dubious use years ago, but since we've got EPG's it's useless as well as distracting and annoying.

Solair

I agree with that, I think RTE is suffering from marketing people gone a bit mad.

There's no need for on-screen logos with EPGs. It is very annoying.

Minstrel27

Mike 1972 wrote: »

If you have an analogue tuner with a auto channel naming feature (works off Teletext signals) RTE2 is still labelled "NET 2" and TG4 is "TNAG"

Its only been a decade (and some) like..........:rolleyes:

Oh right. I thought you were on about some unknown analogue tv system with an EPG.

Sponge Bob

Solair wrote: »

I understand that it might be part of their branding, but normally, diacritical marks i.e. accent marks (fadas) are only used to change the sound of a word when it is being pronounced. Nobody uses them in abbreviations, unless you are reading them phonetically.

Pronounced correctly all the time on The Radio, I cannot speak for the illiterate class of Dub that usually pronounces this as "bored gash" :cool:

icdg

Red Alert wrote: »

The thing I don't see the need for is the on-screen RTE logo. It may have been of dubious use years ago, but since we've got EPG's it's useless as well as distracting and annoying.

RTÉ only adopted DOGs in 2003, well after the digital age had begun. It was for branding reasons far more than to let viewers know what station they were watching. Plus, "Everyone else has one..."

Nowadays the only major non-movie channels that don't sport DOGs are BBC One, BBC Two, and Channel 4. Even they tried them in the early days of digital, viewers in the Nations went ballastic at seeing "BBC ONE Northern Ireland" or such sprawled accross the top of the TV and the BBC hastily abandoned them and haven't used them since on BBC One and BBC Two (though the other BBC channels DO sport DOGs).

As for the RTÉ fada. It is odd to use it in an abbriviation. In fact RTÉ always used "RTE" (no fada) from when it adopted the name in 1966 right up until 1995. The present logo which was adopted in August 1995 is the first logo to read "RTÉ" instead of "RTE". I doubt they would ditch it now - Irish langauge zealots would have a field day. Plus the current RTÉ logo is a design classic - it has been use for fifteen years and shows zero signs of dating. Compare with the 1987 logo which looked dated nearly the day it was unvieled. No wonder it lasted a mere 18 months on screen and was consigned to printed material only for the remainder of its life.

Sponge Bob

Or the original late 1970s yoke They were correct to say eeee when there was no fada, logos being odd like that.

Now there is a fada it is aaaaayy.

Solair

It's not the logo that's the problem, there is no issue with that. The problem is that they've got accents in the channel name on the EPG and elsewhere which is causing display issues.

I've no issue with using it in the logo, but it just does not work in monospace type.

Sponge Bob wrote: »

Pronounced correctly all the time on The Radio, I cannot speak for the illiterate class of Dub that usually pronounces this as "bored gash"

How does one pronounce "RTÉ" ?

"R" "T" "E-fada?"

"R" "T" "Aey"

it's a TLA (Three Letter Acronym)

Sponge Bob

The É in the logo is visually compressed, maybe they should only broadcast it in MHEG as a graphic file not in EIT streams which display a proportional font ...eg bog standard teletext and subtitles

STB

Solair wrote: »

It's not the logo that's the problem, there is no issue with that. The problem is that they've got accents in the channel name on the EPG and elsewhere which is causing display issues.

Emm - what country is your TV set for ? This is a shortcoming in your TVs character set.

I have a three year old combo box that shows the fada's perfectly.

dubhthach

Isn't this not why we have Unicode? I wonder how it would react to the old punc's

eg: ċ ḋ ḟ ġ ṁ ṗ ṡ ṫ

Sponge Bob

the 'buailte' characters = h expressed as a dot over previous character....are part of Unicode.

But I would be happy with ISO-8859-1 support with bog standard fadas and would not worry overly about ISO-8859-14

Anyway what about our own ampersand character ?? The " Tironian Sign ET " That is part of modern standard Irish but not 8859-14 ...just pointing out like.

Solair

Yeah, a lot of people in the 1970s apparently wondered who this really old fashioned telecommunications company called P7T (P Seven T) were.

They were trying to order up things like this :

dubhthach

Sponge Bob wrote: »

the 'buailte' characters = h expressed as a dot ove rprevious character....are part of Unicode.

But I would be happy with ISO-8859-1 support with bog standard fadas and would not worry overly about ISO-8859-14

Anyway what about our own ampersand character ?? The " Tironian Sign ET "

Easy enough to do as it's also in Unicode:
Craic ⁊ Ceol

On my keyboard I just have to type: "\ + &" to get the Tironian ET

The buailte I get by doing "/ + letter". I can also do the following:
ɼ -- long R
ſ -- long s
ẛ -- long s with seimhiu

http://mearchlar.tripod.com/

Sponge Bob

None of me fonts support it

dubhthach

Sponge Bob wrote: »

None of me fonts support it

I take it ye still on WinXP eh?

"Segoe UI Symbol" that ships with Windows 7 has it along with support for all the Ogham characters (Microsoft actually mention ogham support in MSDN)

Couple other fonts would be:
Code2000
Gadelica: http://www.iol.ie/~sob/gadelica/

Rudhraigheacht http://www.folkplanet.com/seanchlo/ruraiocht/

The last one makes reading the 1935 printed version of the Táin more like how it was printed.

Sponge Bob

I think that Teletext in Ogham is just a tad beyond RTÉ's capabilities...or indeed what is expected of them

Then again they are the muppets who are giving us "Irigwe" language codes in the transport stream descriptors if you search around this forum.

fat-tony

Sponge Bob wrote: »

None of me fonts support it

Displaying fine on Firefox on XP;)

watty

dubhthach wrote: »

I take it ye still on WinXP eh?

"Segoe UI Symbol" that ships with Windows 7 has it along with support for all the Ogham characters (Microsoft actually mention ogham support in MSDN)

Couple other fonts would be:
Code2000
Gadelica: http://www.iol.ie/~sob/gadelica/

Rudhraigheacht http://www.folkplanet.com/seanchlo/ruraiocht/

The last one makes reading the 1935 printed version of the Táin more like how it was printed.

All works in my XP.

Including Modern & and 19th century Cyrillic, both sorts Chinese, Modern typographical, printed and ancient Hebrew, Arabic, Korean, etc...

The worst ogham I saw was at the Celtic Park & Gardens off Foynes Rd... Obviously done with an angle grinder.

It's just a matter of installing fonts Spongebob.

I somehow even ended up with a Trains font. Each character is a miniature locomotive or carriage.

cormacl

Based on my checking of the service name details on my dreambox, it's encoded in UTF-8.

I would suspect the receivers/TVs having trouble displaying it however are either only ASCII or Latin1 (Western/ISO-8859-1) compliant. ASCII only systems should show either nothing after the RT or 2 other wild characters.

Here comes the techy stuff..

A latin-1 interpretation of the UTF-8 string RTÉ would be "RTÃ" plus one more wildcard character e.g. a space, underscore etc. An ASCII-only receiver will only show the RT and either gibberish or wildcards for the rest.

The hex sequence I got for RTÉ from the DTT services list was 52 54 c3 89. 0x52 and 0x53 are "RT". The 0xc3 and 0x89 octets are the UTF-8 encoding of É. 0xc3 will be shown as à if interpreted as Latin1 and 0x89 has no Latin1 assignment. ASCII only supports characters to 0x7f. This should explain the formatting I suggested above.

I searched the list of services for similar UTF-8 encodings and found several others such as "C+ ACCIÓN". So RTE are not alone with this behaviour. But indeed we must assume that many other providers just keep things within the ASCII range. Otherwise, we'd surely see plenty of Arabic naming embellishment here and there across services on 13E etc. I only have services from 13/19/28 to check in this regard.

While I was there, I tested an EIT string for a program on TG4 and surprise surprise it was in Latin1 and not UTF-8. So it seems to be inconsistent in this regard. Can the OP confirm how say TG4 EPG program info shows up with the fada because that is in Latin1 from what I can tell.

And on that note, a similar EIT test on a Cyfra+ service showed UTF-8 in use in the EIT for Polish characters. My dreambox seems to render both UTF-8 and Latin-1 EIT data fine so its definitely auto-detecting this. A shame really that this is not just strict UTF-8.

aidan_walsh

Nowadays the only major non-movie channels that don't sport DOGs are BBC One, BBC Two, and Channel 4. Even they tried them in the early days of digital, viewers in the Nations went ballastic at seeing "BBC ONE Northern Ireland" or such sprawled accross the top of the TV and the BBC hastily abandoned them and haven't used them since on BBC One and BBC Two (though the other BBC channels DO sport DOGs).

BBC One HD has one so I wonder if they are testing that for a return, or just using it as a differentiator from BBC One.

watty

Ansi ascii 201 is É
DOS asci 144 is É

195 = 0xC3 = Ã
137 = 08x9 is between 128 and 138, so in Ansi "wraps" to 9 which is TAB
in DOS character set it's ë
195 in DOS character set is a ├ (drawing character)

cormacl

Bum steer on my earlier post.. the UTF-8 I'm seeing on the dreambox is their way of unifying the madness in the specification. So while my box is showing UTF-8 characters to me, it seems they were converted to this by the software and are not transmitted this way originally.

I've looked at the ETSI specs and am assuming that this is the applicable specification in use here..
ETSI EN 300 468
Digital Video Broadcasting (DVB);
Specification for Service Information (SI) in DVB systems
http://www.etsi.org/deliver/etsi_en/300400_300499/300468/01.11.01_60/en_300468v011101p.pdf

In Annex A near the end of the document, it details the selection of character set. The first octet of the string implies or indicates the character set. If this is in the range of 0x20-0xFF, it assumes the default Latin character set they specify which is a slight variant of ISO 6937 where they just added Euro symbol.

ISO 6937 is an odd ball character set that specifies accented characters by using a diacritical character to escape a normal character as a means of specifying the desired accent, be it a grave, acute, dot etc.

So RTÉ will get encoded in hex as 52 54 C2 45. "RT" are the 0x52 an 0x54 octets. Then 0xC2 is the diacritical mark for an acute and 0x45 is "E" which is to be displayed as É. This kind of escaping was common at a time when they wanted to fit the bulk of characters into a single 8-bit range and still accomodate as wide a range of characters as possible. So escaping was the norm.

You can get more details on this encoding here...
http://en.wikipedia.org/wiki/ISO/IEC_6937#Two_byte_characters

If the first character of the string < 0x20, then different character sets can be selected.

e.g...
0x01 ISO/IEC 8859-5 Latin/Cyrillic alphabet
0x02 ISO/IEC 8859-6 Latin/Arabic alphabet
0x03 ISO/IEC 8859-7 Latin/Greek alphabet
0x04 ISO/IEC 8859-8 Latin/Hebrew alphabet
0x05 ISO/IEC 8859-9 Latin alphabet No. 5
0x06 ISO/IEC 8859-10 Latin alphabet No. 6
0x07 ISO/IEC 8859-11 Latin/Thai
0x09 ISO/IEC 8859-13 Latin alphabet No. 7
0x0A ISO/IEC 8859-14 Latin alphabet No. 8
0x0B ISO/IEC 8859-15 Latin alphabet No. 9
0x10 ISO/IEC 8859
...

it goes on and there's even a multi-octet variation of this character set selection.

Flexible for specific charset dsires, but real antiquated and very ambiguous. You could get your É via several of these character sets alone.

TBH.. if this was being done in the last 10 years, they would have selected UTF-8 and be done with it. Unfortunately UTf-8 was only ratified in 1996, far too late to have been more commonly adopted in earlier technologies. Its very dominant now in web, IM and even email use.

Interestingly, the 0xC2 escaping of ISO 6937 has a match to  in the 8859-9 (Latin-5). So it may hint at the receiver/TV defaulting to a specific character set and ignoring the correct default of ISO 6937. But at the OP quite correctly said.. who cares!.. they probably should have ditched the fada for the sake of getting it to render correctly everywhere.