1) "Why Reach for Perfection if 'Meh' is Good Enough? Acceptable Mediocrity in MT and Broader Communication" - Maciej Rączka (memoQ)
We look at the dawn of AI-driven translations with awe and high hopes. Everyone is excited. Any localization professional will say that they reach for the highest quality possible. Few will admit that in some cases, mediocre MT is just fine. We accept low quality communication in many contexts, because it does the job. I believe that the industry will see a much clearer division into high and low quality segments and the customers, users and recipients of some will be just fine with poor, machine-translated websites, products and dialogues. I think that mixed language contents, half-baked translations and unclear communication will persist and accompany many industries. Corporate websites, online shopping, pop media are a few examples of where communication quality matters not as much as we say it should. In my view, we are witnessing a new 'lingua franca', a common pulp language. Machine Translation is the new Latin, which for many centuries persisted, in spite of imperfections, local flavoring and mixed use. I believe that MT will settle for 'meh' in more contexts than expected.
2) "You Need to Calm Down with Your AI Revolution" - Levan Giunashvili (Interlanguage)
Don't worry, we're not going to rain on the tech parade or be nostalgic about the good old days; no "technology bad, typewriter good" vibes here.
But we do want to offer a different perspective, one grounded in reality. High-level executives might sometimes be a tad bit detached from the practical realities on the ground. They may not always know - or want to know - what things look like at the bottom of the pecking order, where real people - linguists, for example - deal with whatever AI-generated gibberish you throw their way. Why should you care if it just saves you money?
Well, there are many reasons why you should care, and one of them is that money you think you're saving. While you might be excited to spend 30% less on your translations today, your overseas sales & marketing department will sing a sadder song tomorrow, and so will you.
Don't let a cloud of buzzwords cloud your mind. Let's discuss the promises of a better bottom line, streamlined processes, and translations served up at no less than lightning speed - and consider the potential costs of getting too carried away by the hype.
1)“如果‘Meh’足够好,为什么还要追求完美?在机器翻译和更广泛的交流中可以接受的平庸”——Maciej Râczka(memoQ)
我们怀着敬畏和厚望看待人工智能驱动的翻译的曙光。大家都很兴奋。任何本地化专业人士都会说,他们尽可能追求最高的质量。很少有人会承认,在某些情况下,平庸的MT就可以了。在许多情况下,我们接受低质量的沟通,因为它能起作用。我相信,该行业将会看到高质量和低质量细分市场的更清晰划分,一些细分市场的客户、用户和接收者将会对糟糕的机器翻译网站、产品和对话感到满意。我觉得混杂的语言内容,半生不熟的翻译,沟通不清,会持续下去,伴随很多行业。公司网站、网上购物、流行媒体是沟通质量没有我们说的那么重要的几个例子。在我看来,我们正在见证一种新的“通用语”,一种普通的低俗语言。机器翻译是新的拉丁语,尽管有不完美的地方风味和混合使用,但它持续了许多世纪。我相信MT会在比预期更多的上下文中满足于‘meh’。
2)“你需要冷静对待你的人工智能革命”——Levan Giunashvili(中介语)
别担心,我们不会给科技游行泼冷水,也不会怀念过去的美好时光;这里没有“技术不好,打字机好”的氛围。
但是我们确实想提供一个不同的视角,一个基于现实的视角。高层管理人员有时可能会有点脱离实际情况。他们可能并不总是知道——或者想知道——处于等级最底层的东西是什么样子,在那里,真实的人——例如语言学家——处理你扔给他们的任何人工智能生成的胡言乱语。你为什么要关心它是否只是为你省钱?
嗯,你应该关心的原因有很多,其中之一就是你认为你正在节省的钱。虽然你可能会为今天在翻译上少花30%而兴奋,但你的海外销售和营销部门明天会唱一首更悲伤的歌,你也会。
不要让一团流行语蒙蔽了你的头脑。让我们讨论一下更好的底线、简化的流程和以不低于闪电般的速度提供翻译的承诺——并考虑一下被炒作冲昏头脑的潜在成本。
以上中文文本为机器翻译,存在不同程度偏差和错误,请理解并参考英文原文阅读。
阅读原文