We're trying to make our "are you a computer or a human?" tests (CAPTCHAs) on our wikis more usable. Specifically, we want to offer the test in lots of languages, not just English. So we need to check whether people can read the non-English images.
We're looking for people who know Bengali, Cherokee, Gujarati, Hindi, Marathi, Sanskrit, Malayalam, Sinhala/Sinhalese, Tamil, and Thai to look at these sample images:
https://www.dropbox.com/sh/i2af7xvn4y593gc/-RRtFyoJji/captchas
* bn Bengali * chr Cherokee * gu Gujarati * hi Hindi (Devanagari script) * mr Marathi (Devanagari script) * sa Sanskrit (Devanagari script) * ml Malayalam * si Sinhala/Sinhalese * ta Tamil * th Thai
This is a followup on Nemo's comment here: https://bugzilla.wikimedia.org/show_bug.cgi?id=5309#c36
Hi Sumana,
I have reviewed Gujarati CAPTCHA images for Gujarati and have provided my comments to Nemo on Mediawiki-i18n list.
There are some problems with spellings in those captcha, where matras are not correctly displayed, which makes it impossible for user to judge the word. Also lets think for once that it is fixed, but 99% users don't have standalone Gujarati keyboard to type. We depend on wiki's typing solutions (e.g. Narayam and ULS) to type. So, the best is to use CAPTCHA in English only on Gujarati wikis.
Regards, Dhaval (User Dsvyas) admin: gu.wp, gu.ws
On Tue, Apr 1, 2014 at 3:33 PM, Sumana Harihareswara sumanah@wikimedia.orgwrote:
We're trying to make our "are you a computer or a human?" tests (CAPTCHAs) on our wikis more usable. Specifically, we want to offer the test in lots of languages, not just English. So we need to check whether people can read the non-English images.
We're looking for people who know Bengali, Cherokee, Gujarati, Hindi, Marathi, Sanskrit, Malayalam, Sinhala/Sinhalese, Tamil, and Thai to look at these sample images:
https://www.dropbox.com/sh/i2af7xvn4y593gc/-RRtFyoJji/captchas
- bn Bengali
- chr Cherokee
- gu Gujarati
- hi Hindi (Devanagari script)
- mr Marathi (Devanagari script)
- sa Sanskrit (Devanagari script)
- ml Malayalam
- si Sinhala/Sinhalese
- ta Tamil
- th Thai
This is a followup on Nemo's comment here: https://bugzilla.wikimedia.org/show_bug.cgi?id=5309#c36
-- Sumana Harihareswara Senior Technical Writer Wikimedia Foundation
Wikitech-ambassadors mailing list Wikitech-ambassadors@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikitech-ambassadors
The french ones look good...
On Tue, Apr 1, 2014 at 10:33 AM, Sumana Harihareswara <sumanah@wikimedia.org
wrote:
We're trying to make our "are you a computer or a human?" tests (CAPTCHAs) on our wikis more usable. Specifically, we want to offer the test in lots of languages, not just English. So we need to check whether people can read the non-English images.
We're looking for people who know Bengali, Cherokee, Gujarati, Hindi, Marathi, Sanskrit, Malayalam, Sinhala/Sinhalese, Tamil, and Thai to look at these sample images:
https://www.dropbox.com/sh/i2af7xvn4y593gc/-RRtFyoJji/captchas
- bn Bengali
- chr Cherokee
- gu Gujarati
- hi Hindi (Devanagari script)
- mr Marathi (Devanagari script)
- sa Sanskrit (Devanagari script)
- ml Malayalam
- si Sinhala/Sinhalese
- ta Tamil
- th Thai
This is a followup on Nemo's comment here: https://bugzilla.wikimedia.org/show_bug.cgi?id=5309#c36
-- Sumana Harihareswara Senior Technical Writer Wikimedia Foundation
Wikitech-ambassadors mailing list Wikitech-ambassadors@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikitech-ambassadors
Swedish looks good apart from one unfortunate combination [1]. In general I would, if possible, avoid combining personal names with sexually transmitted infections.
/André
[1] https://www.dropbox.com/sh/i2af7xvn4y593gc/sIvA9Pg2xu/captchas/sv#lh:null-im...
André Costa | GLAM-tekniker, Wikimedia Sverige | Andre.Costa@wikimedia.se | +46 (0)733-964574
Stöd fri kunskap, bli medlem i Wikimedia Sverige. Läs mer på blimedlem.wikimedia.se
On 1 April 2014 23:13, Taowa Munene-Tardif taowa4@gmail.com wrote:
The french ones look good...
On Tue, Apr 1, 2014 at 10:33 AM, Sumana Harihareswara < sumanah@wikimedia.org> wrote:
We're trying to make our "are you a computer or a human?" tests (CAPTCHAs) on our wikis more usable. Specifically, we want to offer the test in lots of languages, not just English. So we need to check whether people can read the non-English images.
We're looking for people who know Bengali, Cherokee, Gujarati, Hindi, Marathi, Sanskrit, Malayalam, Sinhala/Sinhalese, Tamil, and Thai to look at these sample images:
https://www.dropbox.com/sh/i2af7xvn4y593gc/-RRtFyoJji/captchas
- bn Bengali
- chr Cherokee
- gu Gujarati
- hi Hindi (Devanagari script)
- mr Marathi (Devanagari script)
- sa Sanskrit (Devanagari script)
- ml Malayalam
- si Sinhala/Sinhalese
- ta Tamil
- th Thai
This is a followup on Nemo's comment here: https://bugzilla.wikimedia.org/show_bug.cgi?id=5309#c36
-- Sumana Harihareswara Senior Technical Writer Wikimedia Foundation
Wikitech-ambassadors mailing list Wikitech-ambassadors@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikitech-ambassadors
-- Sincerely
Taowa Munene-Tardif
Wikitech-ambassadors mailing list Wikitech-ambassadors@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikitech-ambassadors
Hi, I checked the Bengali version and it is not displaying properly. I also added a comment on the related bug.
-- *Nasir Khan Saikat* www.nasirkhn.com
On Wed, Apr 2, 2014 at 1:31 PM, André Costa andre.costa@wikimedia.sewrote:
Swedish looks good apart from one unfortunate combination [1]. In general I would, if possible, avoid combining personal names with sexually transmitted infections.
/André
[1] https://www.dropbox.com/sh/i2af7xvn4y593gc/sIvA9Pg2xu/captchas/sv#lh:null-im...
André Costa | GLAM-tekniker, Wikimedia Sverige | Andre.Costa@wikimedia.se| +46 (0)733-964574
Stöd fri kunskap, bli medlem i Wikimedia Sverige. Läs mer på blimedlem.wikimedia.se
On 1 April 2014 23:13, Taowa Munene-Tardif taowa4@gmail.com wrote:
The french ones look good...
On Tue, Apr 1, 2014 at 10:33 AM, Sumana Harihareswara < sumanah@wikimedia.org> wrote:
We're trying to make our "are you a computer or a human?" tests (CAPTCHAs) on our wikis more usable. Specifically, we want to offer the test in lots of languages, not just English. So we need to check whether people can read the non-English images.
We're looking for people who know Bengali, Cherokee, Gujarati, Hindi, Marathi, Sanskrit, Malayalam, Sinhala/Sinhalese, Tamil, and Thai to look at these sample images:
https://www.dropbox.com/sh/i2af7xvn4y593gc/-RRtFyoJji/captchas
- bn Bengali
- chr Cherokee
- gu Gujarati
- hi Hindi (Devanagari script)
- mr Marathi (Devanagari script)
- sa Sanskrit (Devanagari script)
- ml Malayalam
- si Sinhala/Sinhalese
- ta Tamil
- th Thai
This is a followup on Nemo's comment here: https://bugzilla.wikimedia.org/show_bug.cgi?id=5309#c36
-- Sumana Harihareswara Senior Technical Writer Wikimedia Foundation
Wikitech-ambassadors mailing list Wikitech-ambassadors@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikitech-ambassadors
-- Sincerely
Taowa Munene-Tardif
Wikitech-ambassadors mailing list Wikitech-ambassadors@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikitech-ambassadors
Wikitech-ambassadors mailing list Wikitech-ambassadors@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikitech-ambassadors
Even if Romanian was not explictely nominated, I took a look. It seems to me like there is an unusual concentration of potentially offensive words (3/50 images). They're mild, but I would still avoid them.
The images are in question are: https://www.dropbox.com/sh/i2af7xvn4y593gc/IGgbDtSJWS/captchas/ro#lh:null-im... https://www.dropbox.com/sh/i2af7xvn4y593gc/IGgbDtSJWS/captchas/ro#lh:null-im... https://www.dropbox.com/sh/i2af7xvn4y593gc/IGgbDtSJWS/captchas/ro#lh:null-im...
Also, this image does not contain 2 real words (or I can't make them out for some reason): https://www.dropbox.com/sh/i2af7xvn4y593gc/IGgbDtSJWS/captchas/ro#lh:null-im...
This one seems to contain 3 words: https://www.dropbox.com/sh/i2af7xvn4y593gc/IGgbDtSJWS/captchas/ro#lh:null-im...
Some words are also present in more than one image. Is this ok from a security POV?
Strainu
2014-04-01 17:33 GMT+03:00 Sumana Harihareswara sumanah@wikimedia.org:
We're trying to make our "are you a computer or a human?" tests (CAPTCHAs) on our wikis more usable. Specifically, we want to offer the test in lots of languages, not just English. So we need to check whether people can read the non-English images.
We're looking for people who know Bengali, Cherokee, Gujarati, Hindi, Marathi, Sanskrit, Malayalam, Sinhala/Sinhalese, Tamil, and Thai to look at these sample images:
https://www.dropbox.com/sh/i2af7xvn4y593gc/-RRtFyoJji/captchas
- bn Bengali
- chr Cherokee
- gu Gujarati
- hi Hindi (Devanagari script)
- mr Marathi (Devanagari script)
- sa Sanskrit (Devanagari script)
- ml Malayalam
- si Sinhala/Sinhalese
- ta Tamil
- th Thai
This is a followup on Nemo's comment here: https://bugzilla.wikimedia.org/show_bug.cgi?id=5309#c36
-- Sumana Harihareswara Senior Technical Writer Wikimedia Foundation
Wikitech-ambassadors mailing list Wikitech-ambassadors@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikitech-ambassadors
In Catalan, please change this one: https://www.dropbox.com/sh/i2af7xvn4y593gc/AaOp4XTFIx/captchas/ca#lh:null-im... It sounds like joking of Afghan people.
Vicenç
Date: Tue, 1 Apr 2014 10:33:44 -0400 From: sumanah@wikimedia.org To: wikitech-ambassadors@lists.wikimedia.org Subject: [Wikitech-ambassadors] Please help us check these CAPTCHAs for readability
We're trying to make our "are you a computer or a human?" tests (CAPTCHAs) on our wikis more usable. Specifically, we want to offer the test in lots of languages, not just English. So we need to check whether people can read the non-English images.
We're looking for people who know Bengali, Cherokee, Gujarati, Hindi, Marathi, Sanskrit, Malayalam, Sinhala/Sinhalese, Tamil, and Thai to look at these sample images:
https://www.dropbox.com/sh/i2af7xvn4y593gc/-RRtFyoJji/captchas
- bn Bengali
- chr Cherokee
- gu Gujarati
- hi Hindi (Devanagari script)
- mr Marathi (Devanagari script)
- sa Sanskrit (Devanagari script)
- ml Malayalam
- si Sinhala/Sinhalese
- ta Tamil
- th Thai
This is a followup on Nemo's comment here: https://bugzilla.wikimedia.org/show_bug.cgi?id=5309#c36
-- Sumana Harihareswara Senior Technical Writer Wikimedia Foundation
Wikitech-ambassadors mailing list Wikitech-ambassadors@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikitech-ambassadors
On Tue, Apr 1, 2014 at 4:33 PM, Sumana Harihareswara sumanah@wikimedia.orgwrote:
We're trying to make our "are you a computer or a human?" tests (CAPTCHAs) on our wikis more usable.
Croatian (hr) is _completely_ broken. For example, 80% of them (or so) is completely (or half) in Cyrillic. Some older people will be able to read it, but almost nobody in Croatia will be able to enter Cyrillic text, since that is not an official script here.
I will leave this comment on the bug too.
Željko
On Thu, Apr 3, 2014 at 10:46 AM, Željko Filipin zfilipin@wikimedia.orgwrote:
Croatian (hr) is _completely_ broken. For example, 80% of them (or so) is completely (or half) in Cyrillic
Serbian (sr) is strange too. I think both Latin and Cyrillic are official there, but isn't it strange to ask for people to change input method (from Latin to Cyrillic, and vice versa) in the middle of captcha, like here?
https://www.dropbox.com/sh/i2af7xvn4y593gc/ocKv1yBPuf/captchas/sr#lh:null-im...
The above is not the only example.
Željko
Bosnian (ba) has completely Cyrillic CAPTCHAs, as far as I can see, but according to Wikipedia "Standard Bosnian uses a Latin alphabet."[1]
Željko -- 1: https://en.wikipedia.org/wiki/Bosnian_language
wikitech-ambassadors@lists.wikimedia.org