 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 8-10-2012 9:17, Jaime Vives Piqueres wrote:
> On 08/10/12 09:06, Thomas de Groot wrote:
>> Looking forward to that! And take your time. No need to do a hasty
>> job.
>
> My real problem is that I naively decided years ago to offer a
> bilingual site... so the amount of work needed for any update is almost
> twice (the files are shared between both sites, but I need to translate
> the texts, and I'm not very fast at it).
Ah yes, translation is a tricky business if one does not want to fall in
Babel Fishy idioms... ;-)
Thomas
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Thomas de Groot wrote:
> On 8-10-2012 9:17, Jaime Vives Piqueres wrote:
>> My real problem is that I naively decided years ago to offer a
>> bilingual site...
>
> Ah yes, translation is a tricky business if one does not want to fall in
> Babel Fishy idioms... ;-)
although there seems to have been some real progess in the last years,
at least for non-literary texts. For example, here is the lightsys page
translated back to english from the spanish version:
http://translate.google.de/translate?hl=de&sl=es&tl=en&u=http%3A%2F%2Fwww.ignorancia.org%2Fes%2Findex.php%3Fpage%3DLightsys
It's not perfect but quite readable. There must be something
tricky about deciding between I and he in spanish, though ;)
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 8-10-2012 22:47, Christian Froeschlin wrote:
> although there seems to have been some real progess in the last years,
> at least for non-literary texts. For example, here is the lightsys page
> translated back to english from the spanish version:
>
>
http://translate.google.de/translate?hl=de&sl=es&tl=en&u=http%3A%2F%2Fwww.ignorancia.org%2Fes%2Findex.php%3Fpage%3DLightsys
>
>
> It's not perfect but quite readable. There must be something
> tricky about deciding between I and he in spanish, though ;)
I particularly like the /senior intensity/ ;-)
Thomas
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 08/10/12 22:47, Christian Froeschlin wrote:
> although there seems to have been some real progess in the last
> years, at least for non-literary texts. For example, here is the
> lightsys page translated back to english from the spanish version:
>
>
http://translate.google.de/translate?hl=de&sl=es&tl=en&u=http%3A%2F%2Fwww.ignorancia.org%2Fes%2Findex.php%3Fpage%3DLightsys
>
>
>
> It's not perfect but quite readable.
It's better from english to spanish, so I'm pondering if I should
just drop the spanish part of the site.
> There must be something tricky about deciding between I and he in
> spanish, though ;)
It's just that we drop pronouns, and this is precisely where automated
translation fails most: deducing from context.
--
Jaime
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 09/10/12 09:10, Thomas de Groot wrote:
> I particularly like the /senior intensity/ ;-)
>
The logic behind that bad translation is particularly twisted... here
it got excessively optimistic and creative, and translated "alto rango"
as "senior" instead just using the literal (and correct) translation:
"high range". In spanish, "alto rango" is used traditionally to refer to
the same meaning as "senior". Curiously, I think Google translator got
there because its dictionary is missing the additional meaning of
"rango" as "range".
--
Jaime
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 9-10-2012 10:36, Jaime Vives Piqueres wrote:
> The logic behind that bad translation is particularly twisted... here
> it got excessively optimistic and creative, and translated "alto rango"
> as "senior" instead just using the literal (and correct) translation:
> "high range". In spanish, "alto rango" is used traditionally to refer to
> the same meaning as "senior". Curiously, I think Google translator got
> there because its dictionary is missing the additional meaning of
> "rango" as "range".
Nice example indeed. In the end, in addition to syntactic details, it is
the dictionary which should be as filled as possible, but with the
additional problem of getting the right choice of words.
Thomas
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Here is another rendering. The .gray value of each non-gamut-mapped color was
taken; the color was desaturated; and then the value was adjusted to the
original .gray level. Finally, the spectrum was normalized as a whole.
Post a reply to this message
Attachments:
Download 'spectrum_by_gray.png' (7 KB)
Preview of image 'spectrum_by_gray.png'
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Am 21.10.2012 20:24, schrieb Cousin Ricky:
> Here is another rendering. The .gray value of each non-gamut-mapped color was
> taken; the color was desaturated; and then the value was adjusted to the
> original .gray level. Finally, the spectrum was normalized as a whole.
>
The common way for *visual* best results works like this:
transform the rgb out-of-gamut value to L*a*b and from there to Lch.
Lower the chrominance channel until the rgb value falls back into gamut.
This is the tricky part as you do not know before-head when this happens
so some clever interpolation method is needed.
-Ive
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Ive <ive### [at] lilysoft org> wrote:
> The common way for *visual* best results works like this:
> transform the rgb out-of-gamut value to L*a*b and from there to Lch.
> Lower the chrominance channel until the rgb value falls back into gamut.
> This is the tricky part as you do not know before-head when this happens
> so some clever interpolation method is needed.
I'll have to do a little more reading and a lot more grokking before I can do
this.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Am 22.10.2012 20:49, schrieb Cousin Ricky:
> Ive <ive### [at] lilysoftorg> wrote:
>> The common way for *visual* best results works like this:
>> transform the rgb out-of-gamut value to L*a*b and from there to Lch.
>> Lower the chrominance channel until the rgb value falls back into gamut.
>> This is the tricky part as you do not know before-head when this happens
>> so some clever interpolation method is needed.
>
> I'll have to do a little more reading and a lot more grokking before I can do
> this.
>
>
rgb -> Lch (and visa versa) conversion is part of CIE.inc and for
finding the best Lch chroma value something like a binary search should
do the trick: reduce the actual value by half, convert back to rgb and
check if it is inside or outside the gamut. If still outside reduce
chroma again by half, if inside use the middle between the previous and
current value. Repeat this until a certain threshold is reached.
I'll have to admit that I was lazy and didn't implement it but this is
how the ICC recommends gamut mapping for the perceptual viewing intend.
-Ive
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
