jump to navigation

A new kind of spam comment? March 20, 2012

Posted by Jill (@bonnjill) in Random musings.
trackback

This is a question for all my fellow bloggers. I was going through my “spam” comments today and found a couple that I would normally deem to be spam since they follow a pattern of deliberately misspelling a word in the first sentence. However, in reading the comment I found someone took the time to actually write about translation. Here are some examples:

Comment on my “Wie gut ist dein Deutsch” TGIF video:

Great idea. Start with a simple situloon and work from there. As a client I don’t want heaven I just want what you offered in your newsletter. My biggest problem is that translators translate the same glossary word across different texts and come up with different translations for the same terms and words. This gets worse if multiple translators are involved. Especially when creating software and help files and FAQs this is a huge problem: Menu items always have to have the same translation for example, as users can otherwise not understand how the text ralates to their software.There is no grading involved. There simply is only ONE translation and that’s it. I don’t know why an early version or any version would need a grading system. If I as a client want a translation of a word or term in the glossary fixed, so that it always is translated in the same way, then that’s what I want. That’s why I place it in the glossary. I do this to take away ANY options from the translators to get the translation wrong.If you can deliver what you are suggesting in the newsletter, you would provide a great situloon. From the comments I can see above it appears you have translators responding who want a grat Glossry for entirely different reasons than suggested by you. Stick with what you suggested, and you will make many of your clients happy who want to get better quality and more consistent translations across multiple documents.I do NOT want a glossary such as trainer -> (sport) istruttore, allenatore, trainer, mister That already exists in translation software. There is no need to reinvent this.I want a glossary that takes away options from translators and forces them to always translateX with Y.This is what I understand you are planning and this would be a great situloon.

Second comment in reply to Lisa Davey’s comment:

I don’t think there are all that many comments out there dnaparsgiig MT across the board, if you take a closer look. I’m a professional translator but I certainly see the value in having MT tools available for people to get the quick gist of a web page, for instance. Most of the my god, look at this terrible MT output commentary you see (at least the stuff that’s worth reading at all) will be criticizing inappropriate uses of MT, not MT per se—cases where it’s used to produce absurd signs at the Beijing Olympics, or gibberish on a website that’s actually meant to market services or products to readers of another language. Google Translate is a fine and helpful tool, and I turn to it myself when I want to get a dim idea of what some Russian or Korean just wrote. I don’t write a snarky blog post about Google Translate unless there’s a company using its output, unedited, to try to sell web services to the Japanese market or the like. And then the focus of my post is going to be This company hasn’t got a clue, not Google Translate sucks, ha ha.

Comment on my “Translators do it better” post:

Very early on, the members of Vox Clara agered to something they eventually called the Moroney Principle. The principle maintains that wherever possible the current ICEL translations spoken by the people should be retained provided they are not too distant from what LA proposes. Yet, if the literal route was chosen it would have given parishes all the more reason to either sing the readings, or if they were unable to sing the readings perhaps they could sing the acclamations. When the Latin text is employed the way the different responses to Verbum Domini are registered most effectively in the mind depends upon how Verbum Domini is sung. The sung text cues the different response. Thus, not only could you have the literal translation, but in order for there to be a proper pastoral cue it would necessitate singing the acclamations. How they are sung of course would not have to be what the latin chants propose, but could be adapted for the English Language context. Fr. Joncas might at least agree with that goal. Maybe for the sake of singing they should be translated literally. I am sure that Fr. Joncas would agree that singing an acclamation heightens the effect of it.

They appear to be coming from some Facebook profiles. What say you, fellow bloggers? Spam or not spam? This is hard to determine!

Advertisements

Comments»

1. Margaret - March 21, 2012

I haven’t had such long ones, but I have had ones based on religious weblogs. I find one hit for ‘Very early on, the members of Vox Clara agreed’, i.e. with the correct spelling of the last word:

http://www.praytellblog.com/index.php/2010/11/04/empathy-for-the-translators-of-the-ordo-missae-1-the-dialogues-of-the-liturgy-of-the-word/

Someone has automatically collected texts from the Web – why religious blogs? – and misspelled odd words. I think the misspelling in spam, as you presumably think too, is so that by Googling the spammers can later see where their spam successfully appeared.

The recent comments that have been annoying me have been these: I have imposed captcha for all comments (is there a better way?), so they don’t appear until approved by me. But there are obviously programs out there that can automatically read my captchas by OCR. So I keep having to delete comments waiting for approval.

2. Margaret - March 21, 2012

Actually, religious weblogs probably often refer to translations of the Bible, so the link is translation rather than religion.

3. Jeff Hershberger - March 21, 2012

I’ve also gotten comments where I couldn’t quite tell whether or not they were spam. None of mine were this long, though.

It used to be easy to tell a real comment from spam: the spam comments all had links in them, and the real comments actually said something on point. You delete the one and leave an enthusiastic reply to the other. The problem is, some real human commenters aren’t that great at making their point – sure, the bots are sounding smarter with every passing year, but also some people are just lousy at expressing themselves. So for the ones in the gray area, I use a sort of “does it make sense” heuristic to decide whether or not to delete something: if the comment is harmless and appears to be addressing the subject matter, I leave it (but I don’t bother to answer it).

The motivations for posting a spam comment can be a little inscrutable. Anything with a link in it is either trying to sell people something on click-through or trying to build the google page-ranking of the target page. The text with the link in it can be generated automatically, and the algorithms can be clever enough to choose text that’s almost on-topic. But a comment without a link … why? Is the commenter’s profile page the link they’re trying to push? Or is there a link on the profile page that they’re trying to drive people to? I don’t know.

4. Jill (@bonnjill) - March 21, 2012

Thanks for the feedback, Margaret and Jeff. I erred on the side of caution and have deleted them from the ‘Pending’ folder, because, as Jeff said, it ultimately didn’t make sense.

Margaret - March 21, 2012

They are definitely spam – sorry if I didn’t make it clear. I forgot to say that the comments I have been getting are usually allegedly via Facebook pages too. I assume that the whole thing is done automatically so they post to tens of thousands of websites at once, and the program they are using automatically creates or borrows a Facebook address.

5. Alison - March 21, 2012

I think it must be something along these lines:

http://en.wikipedia.org/wiki/Blog_scraping

I remember reading about it somewhere, probably on another translation blog, but can’t find the reference now.

(I know I haven’t posted on here before and I’ve included a link but I am a human and not a spambot, I promise…!)

Jill (@bonnjill) - March 21, 2012

You’re absolutely right! I googled several of the comments and they led to various blog posts. The commenters had nothing to do with the blog posts though, and their e-mail addresses did not match up with their names. The links led to Facebook profiles that appeared to be attempts at phishing. Thanks, everyone! I have learned these comments should just be marked as spam and deleted. You learn something new every day…

6. Leon Hunter (@lhunterb) - March 21, 2012

Spam them! In my grat Glossry, these dnaparsgiig comments most definatly kwalify as spam 🙂

7. Judy Jenner (@language_news) - March 22, 2012

I’d say spam, yes. These are much longer than anything we’ve ever seen in our spam folders. Odd. Spammers work in mysterious ways, don’t they? 🙂

8. Catherine Jan (@translatetrad) - March 27, 2012

Jill, if I were you, I wouldn’t subject my readers to this kind of situloon.

I set up my anti-spam plug-in to trash any comments containing: beatiful, luv, creativ, awsome and yur. (Why are so many spammers against vowels?) I also don’t tolerate four or more exclamation points in a row.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: