原站英文机器转录 + 本地中文机器初译(未完成逐句听审) 原作者及许可见 attribution.txt;中文由九天新增,不代表原作者认可。 00:00:00.000 EN: Good. So this is the lecture on Semantic Web Knowledge Engineering. My name is Markus Stocker. Hello, everyone. 中文初译: 很好,这是关于语义网络知识工程的讲座 我叫马库斯史托克 大家好 00:00:13.740 EN: I am filling in for Søren today, because Søren had another meeting. This could happen a few times during the semester, so you might see me or one of our colleagues. 中文初译: 今天我替瑟伦接班,因为索伦又开了一次会议。这可能会在学期发生几次,这样你就可以看到我或我们的同事之一。 00:00:26.960 EN: And you had, as far as I'm told, the first lecture last week. And this was largely an introduction to the semantic web. 中文初译: 据我所知,你上星期第一场演讲 基本上是对语义网页的介绍 00:00:39.380 EN: And today, we are diving a little bit deeper into the topic of semantic web technologies, in particular, starting with the RDF data model, the research description framework data model. 中文初译: 今天,我们正在深入探讨语义网络技术, 特别是从“RDF”数据模型开始, 研究描述框架数据模型。 00:00:54.700 EN: So I will tell you a little bit more about this. And let's see how much time we have. Maybe we can go also to the third chapter of this class. 中文初译: 因此,我要告诉你们更多一些关于这个问题。让我们看看我们有多少时间。也许我们可以也去这个班的第三堂课。 00:01:08.360 EN: So motivation. So RDF and semantic web technologies is largely technology to encode meaning, meaning of natural language text, of terms. 中文初译: 所以动机。所以“RDF”和语义网络技术 主要是用来编码术语的含义, 自然语言文字的意义。 00:01:23.940 EN: So what we start here with the motivation is sort of a sentence, a natural language sentence, a piece of knowledge that the theory of relativity was discovered by Albert Einstein. 中文初译: 因此,我们从这里开始的动机是 一种句子,一种自然的语言句子, 一个知识的一部分, 相对论的理论 是阿尔伯特·爱因斯坦发现的。 00:01:38.260 EN: So most of you probably are familiar with XML technology. And this slide proposes that you can actually encode this piece of knowledge here in various ways in XML. 中文初译: 所以你们大部分人可能都熟悉“XML”技术。这个幻灯片建议, 您可以在“ XML” 中以不同的方式将这一知识编码成“ XML ” 。 00:01:50.420 EN: So you could start with the theory and say that there is a theory here with the element theory in XML that has a name. 中文初译: 所以,你可以从理论开始 说这里有一个理论 用"XML"中元素理论的理论 它有名字。 00:02:00.400 EN: So this is the perspective of the theory, has a name, theory of relativity. And of course, the theory has a discoverer. 中文初译: 这就是理论的视角,有名称,相对论。当然,理论有一个发现者。 00:02:07.160 EN: That will be, in this case, Albert Einstein. This is one way to encode this knowledge in XML. But you could also encode it differently from the perspective of the person. 中文初译: 这将是阿尔伯特·爱因斯坦(Albert Einstein)。这是将这种知识编码为“XML”的一种方式。但您也可以从个人的角度将其编码不同。 00:02:22.300 EN: So you could say, OK, we have a person here that has a name that will be Albert Einstein. And this person has discovered some theory. 中文初译: 所以,你可以说,好吧,我们这里有一个人 名字是艾伯特·爱因斯坦。这个人发现了一些理论。 00:02:31.080 EN: That will be the theory of relativity. So you can see here two different ways of encoding, especially the same information. 中文初译: 这将是相对论理论。所以你可以看到两种不同的编码方式, 尤其是相同的信息。 00:02:39.740 EN: But of course, this has a repercussion on systems because they have to sort of parse and be aware that the sorting, the ordering, the perspective is different. 中文初译: 当然,这对系统有影响 因为它们必须进行分析 并意识到分类、顺序、观点是不同的。 00:02:51.880 EN: You could have yet another one, person, again, the name of the perspective of the person. But then say, OK, instead of having a child element name of the person, we could have encode the name as an attribute here, Albert Einstein. 中文初译: 你可以有另一个名字,一个人,再次, 个人的角度的名称。但是,然后说,好吧,我们也许可以把这个名字编码成一个属性, Albert Einstein。 00:03:13.940 EN: And then the theory that he discovered, theory of relativity as a child element. So these are three different ways. 中文初译: 然后他发现了一种理论, 相对论作为儿童元素。所以这三种方式是不同的。 00:03:23.920 EN: And sure, you could come up with four at least. You could do the same with the theory, name of the theory as an attribute, and probably even more. 中文初译: 当然,你至少可以拿出四个。你可以用理论来做同样的事情, 将理论命名为属性,甚至可能更多。 00:03:35.440 EN: So there is no unique way in XML to represent knowledge. And if you do this, or the lack of such a unique way has some problems or makes it challenging to integrate this data. 中文初译: 所以在“ XML ” ( XML) 中没有代表知识的独特方式。 如果你这样做, 或者说缺乏这种独特的方法, 就会有一些问题或者让整合这些数据成为挑战。 00:03:53.600 EN: There's a why there. I'm not sure. Maybe we can pause here for a few minutes. If you have an answer to this, you can either speak up or write in the chat or take it as a homework also. 中文初译: 也许我们可以在这里暂停几分钟 如果你对此有答案 你可以大声说出来 或者在聊天中写作 也可以当做家庭作业 00:04:09.600 EN: But I'm giving a minute to see if anyone has an answer to why this is hard to integrate. And you can speak up or use the chat. 中文初译: 但我要花一点时间 看看是否有人能回答为什么很难融合。你可以大声说话或者用聊天方式。 00:04:24.980 EN: So I see Luca suggests there is no uniform structure. And that's correct. So there are different structures. And of course, consuming clients that use this data need to then be aware of these different structures and also need to be programmed against these different structures. 中文初译: 我看到卢卡说没有统一的结构。没错,所以有不同的结构。 当然,消费使用这些数据的客户 需要意识到这些不同的结构, 也需要根据这些不同的结构来编程。 00:04:52.360 EN: Because as you know, there is a sorting in XML. So theory name is not same as a person name. So you have to accordingly parse these documents and read them accordingly. 中文初译: 因为如您所知, 在“ XML” 中有排序。 所以理论名称与个人名称不同。 因此您必须相应地分析这些文件并阅读它们 。 00:05:10.300 EN: So that's a challenge then for all the consuming agents that have to be aware of this different structure. And then we have also Hanan, based on person who defines a so different way to represent it. 中文初译: 因此,对于所有消费代理商来说,这是个挑战, 他们必须意识到这个不同的结构。然后我们还有Hanan, 依据的就是那些定义了一种不同方式来代表它的人。 00:05:38.240 EN: I'm not entirely sure how to parse this. But yeah, so you have maybe these different perspectives. So you have on one hand the perspective of the theories. 中文初译: 我不完全确定该如何分析。但是是的,所以你也许有这些不同的观点。 所以你一方面有理论的观点。 00:05:48.740 EN: And you can imagine having here an XML document with many theories. So you have a list of theories or a list of person. 中文初译: 你可以想象这里有一份"XML"文件 有很多理论。所以你有一份理论清单或个人名单。 00:05:56.460 EN: And that's, of course, a very different representation on your data. And again, accordingly, you need to develop systems to have either a perspective on theories, so the list of theories, and then processing theories, or a list of person and processing persons. 中文初译: 当然,这是数据上非常不同的表达方式。同样,你需要开发一个系统 来对理论有观点, 比如理论清单,然后处理理论, 或者个人和加工人员的名单。 00:06:20.960 EN: So this has also implications on the system probably. So if you want to develop a catalog of theories, you would rather probably have theory element and then have them as discover or the discoverer as an element of theory. 中文初译: 这对系统也有影响。所以如果你想要开发一个理论目录, 你可能宁愿有理论元素, 然后把它们作为发现者或发现者作为理论元素。 00:06:39.680 EN: If you want to have a catalog, if your system is a catalog of people, then you might probably choose the perspective of the people, list of persons. 中文初译: 如果你想要一份目录, 如果你的系统是人口目录, 那么你可能会选择人们的观点, 个人名单。 00:06:49.360 EN: And then it happens that some people are discoverer of a theory and some are not, and so on. So this is a choice probably that affects or depends only also on the system that you are going to develop and the purpose that that system supports. 中文初译: 然后发生的情况是,有些人是理论的发现者,有些人不是,等等。所以这个选择可能影响或只取决于你将要开发的系统 以及该系统支持的目的。 00:07:15.420 EN: And accordingly, you model the data. Good. A few thoughts on this. You can take it also as a homework. Think a little bit further, what are the implications of this different modeling. 中文初译: 因此,您可以模拟数据。很好。一些想法。你可以把它也当作家庭作业。 再仔细想想,这个不同的模型会有什么影响。 00:07:31.980 EN: And sort of the last line here suggests that RDF helps to solve this problem. And I'm leaving this open at this point, maybe hopefully at the end of the lecture, we have some answers to this. 中文初译: 这里最后一句话暗示, “RDF” 有助于解决这个问题。我暂时不讨论这个问题, 也许在演讲结束时, 我们对此有一些答案。 00:07:44.280 EN: And perhaps also not just with RDF, but in future lectures, you will also hear about the RDF schema. And this is yet another piece of technology that helps with this problem. 中文初译: 也许不仅有“RDF ” ( RDF ) , 而且在未来的演讲中,你也会听到“RDF ” ( RDF ) 的假想。 这是又一项有助于解决这个问题的技术。 00:07:57.120 EN: Good. So the goals today are to understand the RDF data model. We will talk about what are URIs, what are IRIs. 中文初译: 很好,今天的目标是了解“RDF”数据模型。我们将讨论什么是“URIs”,什么是“IRIs”。 00:08:10.980 EN: It's still morning. And then we have triples as an important concept in the RDF model. Resources, we will talk about literals, blank notes, and also lists, containers as another feature of RDF. 中文初译: 现在还清晨。然后我们有了三重概念作为“RDF”模型中的重要概念。资源,我们将讨论字面、空白笔记以及清单,容器作为“RDF”的另一个特征。 00:08:34.410 EN: Prerequisites for this class are to have a basic understanding of web technologies. I'm sure that that's given here. 中文初译: 这个班的前提条件是 基本了解网络技术。我确信这里已经给出了这一点。 00:08:43.570 EN: And most of you, certainly, or all of you know basic web technologies. Like, for instance, markup language is HTML come to mind. 中文初译: 当然,你们大部分人或者你们所有人都知道基本的网络技术。比如说,标记语言就是HTML。 00:08:53.190 EN: But also XML, that's all useful. And also an understanding of what the data type is is useful. And again, most of you have certainly already touched programming languages. 中文初译: 但“ XML” 也是有用的。 了解数据类型是有用的。 另外, 你们中的大多数人肯定已经接触到了编程语言 。 00:09:07.790 EN: So you're certainly familiar with what the data type is. So RDF. RDF is an acronym. Stands for the Resource Description Framework. 中文初译: 所以,你肯定熟悉数据类型。所以“RDF”是缩写。“RDF”是资源描述框架的缩写。 00:09:19.300 EN: It has been a recommendation by the W3C for quite some time. So over 20 years now, in fact, these technologies are even a little bit older, at least in the conceptual phase. 中文初译: 这是W3C建议了相当长一段时间。 事实上,20多年来,这些技术甚至比现在更老一些,至少在概念阶段是如此。 00:09:34.480 EN: So I think the semantic web and some of the basic technologies were already sort of conceived by Tim Berners-Lee and some colleagues in the early 90s. 中文初译: 因此我认为语义网和一些基本技术 已经由蒂姆·伯纳斯-李 和90年代初的一些同事 构思。 00:09:45.100 EN: So they are almost as old as the web itself, which is quite natural because I think Tim Berners-Lee and his colleagues, they were aware that in order to encode or make the data published on the web useful for machines, it's certainly useful to not just provide layouting, so styling and formatting of text so that the system knows how to display data on screens. 中文初译: 所以它们几乎和网络本身一样老, 这是很自然的, 因为我认为Tim Berners-Lee和他的同事, 他们意识到,为了编码或使网上公布的数据对机器有用, 不仅提供布局, 00:10:23.680 EN: But also, in order to make this data useful that is published on websites, it's important for machines to have access to the meaning of the data. 中文初译: 但是,为了使这些在网站上公布的数据有用, 机器必须能够了解数据的含义。 00:10:34.440 EN: And therefore, so this idea was pretty clear early on. In the early 90s, I would say already. And then the technologies for doing this, for encoding semantics of data in a formal way, in a machine-readable way, came then later towards the end of the 90s, as you see. 中文初译: 因此,这个想法在早期就非常清楚了。在90年代初期,我想已经说过了。然后用一种正式的方式,以机器可读的方式, 将数据词义编码的技术, 在90年代的末期, 正如你所看到的。 00:10:56.400 EN: And then also early 2000s, we had the RDF schema. And you will hear a little bit more about this. And then later on, also, the web ontology language as for the technologies of the semantic web stack. 中文初译: 还有2000年代初期,我们有了“RDF”的图案。你会听到更多一些关于这一点的信息。 稍后,还有网络文学语言,比如语义网页技术。 00:11:11.260 EN: And then, so 1998, the first recommendation. And then later on, quite a bit later, we had version 1.1, 2014. And to my, I haven't checked, but I think it's still at version 1.1. 中文初译: 然后,1998年,第一个建议。后来,相当晚一点,我们有了1.1版,2014年。 对我而言,我还没有检查过,但我想它仍然在1.1版。 00:11:28.420 EN: So RDF is a data model. So similar like other data models. So for instance, the relational data model that you know from relational database management systems, RDF is also a data model. 中文初译: 所以“RDF”是一个数据模型。与其他数据模型相似。例如,从关系数据库管理系统中你所知道的关联数据模型“RDF”也是一个数据模型。 00:11:45.050 EN: And originally, it was used for metadata. So the original intent was to use RDF to represent metadata about web resources, so HTML websites, but also linked images, linked videos. 中文初译: 最初,它被用于元数据。所以最初的用意是使用“RDF”来代表有关网络资源的元数据,例如 HTML 网站,但是连线图像,链接视频。 00:12:01.680 EN: So basically, digital resources accessible on the web. But then the model was quickly generalized, since you can encode, really, information about any resource. 中文初译: 基本上,数字资源可以在网络上查阅。但是这个模型很快被普及了, 因为你可以编码任何资源的信息。 00:12:14.100 EN: So also conceptual resources, also physical resources like people or samples. You can use this data model to encode information about resources beyond web resources. That was also pretty clear early on, and it was then fast extended to these other entities. 中文初译: 比如概念资源,还有人或样本等物质资源。你可以使用这个数据模型来编码网络资源以外的资源信息。这也在早期就非常清楚了,然后迅速推广到这些其他实体。 00:12:40.940 EN: And it essentially supports to encode structured information, and we will see how this works. And it's often used as a universal, certainly machine-readable exchange format for data. 中文初译: 它基本上支持编码结构化信息, 我们将看到它是如何运作的。它经常被用作一个通用的,当然是机器可读的数据交换格式。 00:12:59.700 EN: Like other formats, JSON is also very common. XML is still also quite common. And the key difference, I guess, between RDF and other data models is that RDF is a so-called graph data model. 中文初译: 与其他格式一样,JSON也非常常见。“XML”仍然相当常见。我想,“RDF”和其他数据模型之间的关键区别是“RDF”是一个所谓的图形数据模型。 00:13:22.720 EN: So the data is structured in graphs, and I will show you a little bit more about this. So we have, in particular, vertices and edges, so nodes and directed edges in the RDF data model. 中文初译: 因此数据结构以图表为结构, 我会向大家展示更多一些关于这个的。 所以我们在“ RDF” 数据模型中, 特别有脊椎和边缘, 所以节点和定向边缘。 00:13:41.820 EN: So the parts of the RDF graph are, to remember, on one hand, URIs. URIs are the uniform resource identifiers, and they are used to reference resources unambiguously. 中文初译: 所以“RDF”图中的部分, 一方面是“URIs”。“URIs”是统一的资源识别符, 它们被用来毫不含糊地参考资源。 00:13:59.020 EN: So they are global identifiers, so they are not local identifiers like in databases, like some numbers. But they are, on the web, globally identified. 中文初译: 所以它们是全球性的识别数据,所以它们不像数据库里那样是本地的识别数据, 像一些数字一样。但是在网络上,它们在全球被识别。 00:14:13.120 EN: And this type of identification system is used in RDF to identify resources unambiguously. So we have URIs, which is an important concept. 中文初译: 这种身份识别系统在“RDF”中被用来明确确定资源。因此我们有了“URIs ” , 这是一个重要的概念。 00:14:27.260 EN: Then we have literals. Literals are used to describe data values. And data values usually don't really have a need to be identified, so 314 or, here, 100 kilometers an hour. 中文初译: 然后我们有了字面文字。 字面文字用来描述数据值。 而数据值通常并不真正需要识别, 所以314或者, 这里, 每小时100公里。 00:14:45.920 EN: And these are things that don't have an own identity, and therefore, they are not assigned a URI. And yeah, they are data values. 中文初译: 这些是没有自己身份的东西, 因此,它们没有被分配到“ URI ” 。 是的,它们是数据值。 00:14:56.620 EN: And in the RDF data model, they are called literals. And then, in addition, we have blank notes. And blank notes, they are used, they are like sort of regular notes or regular resources, except that they are not named, so they are not identified. 中文初译: 在“RDF”数据模型中,它们被称为字面文字。此外,我们还有空白的笔记。这些笔记被使用,它们就像普通笔记或经常资源一样,除了没有命名,所以没有被识别出来。 00:15:19.540 EN: And they are used, as I will show in a later slide, as a technique to describe more complex resources by introducing these so-called blank notes, which are just helper notes without any identification. 中文初译: 正如我将在以后的幻灯片中显示的那样,它们被用作一种技术,通过介绍这些所谓的空白笔记来描述更复杂的资源,这些笔记只是帮助人笔记,没有任何标识。 00:15:38.440 EN: And this will be clear in a later slide. So let's look at an RDF graph. Can I ask a question here? Yes. Where do you want to go? 中文初译: 稍后的幻灯片中将清楚说明这一点。让我们看看一个“ RDF” 图。 我能在这里问个问题吗?是的。你想去哪里? 00:15:50.940 EN: So like the blank notes, we also have the data quality issues when we are saying that we have empathy or blank notes. 中文初译: 和空白笔记一样,我们也有数据质量问题 当我们说我们有同情心或空白笔记时。 00:16:03.240 EN: So how is it different from the blank notes that we have in the data sets as compared to the blank notes that we are defining in the RDF? In which data sets, you mean? 中文初译: 那么,与我们在“RDF”中定义的“空白笔记”相比,我们数据集中的空白笔记又有什么不同呢? 你是说在哪些数据集中? 00:16:17.260 EN: Like, usually, we have some missing values. We don't know either. So they are also called treat as a, and that is a very big issue for when we are predicting things. 中文初译: 通常,我们有一些缺失的值。我们也不清楚。所以它们也被称为“一个对待”, 而当我们预测事情的时候,这是一个非常大的问题。 00:16:29.580 EN: Right. Yeah. Yeah, so this is a little bit different. And maybe it becomes clear later. But so missing data, missing values, so you have a classical data sheet, an Excel sheet or so, and you have an NAN or even an empty cell somewhere. 中文初译: 对,是的,所以这有点不同。也许以后就会清楚了。但是缺少数据、缺失值, 所以你有一个古典的数据表, Excel的表格等等, 你有一个NAN甚至空的单元格。 00:16:50.880 EN: That's a different blank note, let's say, than in the RDF, a blank note. So it would be clear what we mean by RDF blank notes. 中文初译: 换句话说,这是个不同的空白字, 比在“RDF”里写的“RDF”一个空白字。所以我们用“RDF”的空白字词是什么意思就很清楚了。 00:17:02.240 EN: But they are not empty, let's say. So they are indeed notes in the graph. They exist. They can be traversed. The system can go from one note to over an edge to a blank note and over its edges to other notes. 中文初译: 但它们不是空的, 可以说。 因此它们确实是图表中的笔记。 它们存在。 它们可以被穿行。 系统可以从一个笔记到一个边上, 到一个空白的笔记, 然后从它的边缘到其他笔记。 00:17:20.400 EN: So they do exist. And they are sort of actionable. And a data set empty field is, yeah, empty, doesn't have a value. 中文初译: 所以它们确实存在。 它们可以操作。 而一个数据集空字段是, 是的, 空的, 没有值 。 00:17:35.830 EN: And anyway, it's a different notion also because the data set is a holder of data. And if there is a missing value somewhere, the interpretation of this is, as you said, that simply the value is not known, NAN typically. 中文初译: 无论如何,这是一个不同的概念 也因为数据集是数据持有者。如果在某处缺少一个价值的话, 正如你说的,对这个价值的解释是,仅仅该价值并不知道, 通常是NAN。 00:17:55.320 EN: Or even if it's empty at all, you don't even know if it was considered or left out by the input, if it was an error or so. 中文初译: 即使它是空的,你甚至不知道它是否被考虑过 或被输入遗漏了, 它是否是一个错误。 00:18:04.860 EN: So that's a completely different notion and concept here. And yeah, they should be mixed. But I hope it will be clear maybe what an RDF blank note means then later on in the lecture. 中文初译: 因此,这是一个完全不同的概念和概念。是的,它们应该混杂在一起。但是我希望它会清楚,也许“RDF”的空白字条意味着什么, 稍后在演讲中。 00:18:25.200 EN: Yeah, thank you. You're welcome. Any other question? Just the chat doesn't have anything. Good, then I will just move on with the example. 中文初译: 谢谢,别客气 还有什么问题吗? 00:18:39.460 EN: And so as you can see here graphically, so this is a small RDF graph. And you can see here we have some notes. So there are two types of notes here, the green ones and the red ones. 中文初译: 这样你可以看到这里的图形, 因此这是一个小的“ RDF” 图。 您可以在这里看到我们有一些笔记。 所以这里有两种笔记, 绿色和红色。 00:18:57.360 EN: They're also rectangular and overall notes. So this has a meaning. So there are two different types of notes. We were talking about URI-identified resources. Those are the green notes. 中文初译: 它们也是长方形的和整体的笔记。 所以这有其含义。 所以有两种不同的笔记。 我们谈论的是“ URI” 识别的资源, 这些是绿色笔记。 00:19:12.970 EN: They have an identity. Leipzig is a city with an identity. Burkhard Jung is a person with an identity. The Social Democratic Party is an institution, an organization, or a party with an identification. 中文初译: 他们有身份,Leipzig是一个具有身份的城市,Burkhard Jung是具有身份的人,社会民主党是一个机构、组织或具有身份的政党。 00:19:31.780 EN: And so has Saxony as a state in Germany and Germany as a country. They're all entities with an identity. So you can identify this by URIs. 中文初译: 萨克森州作为德国和德国的一个国家也是一样。它们都是有身份的实体。所以你可以用“URIs”来识别这一点。 00:19:45.420 EN: And this is not clearly visible here in this graph. It's a little bit abbreviated. This is a sort of a local name, Leipzig. 中文初译: 这张图中无法清楚地看出这一点。 它略微缩略了一点。 这是一个本地名称, Leipzig 。 00:19:53.260 EN: But you can understand this, and I will mention this later also, as a local name in a namespace. That would be then a full URI. 中文初译: 但是,你可以理解这一点,我稍后也将提到这一点,作为一个地名空间的本地名称。那将是一个完整的“URI”。 00:20:04.690 EN: But then we have also this type of notes here. And these are these literals, which I just mentioned. So these are data values. They don't have an identity. 中文初译: 但是我们在这里也有这样的笔记。 这些是字面的,我刚才提到的。所以这些是数据值。它们没有身份。 00:20:16.000 EN: So they are not identified with a URI. They are just data values, 12.3833. And we also have attributes, properties, or edges in the graph. 中文初译: 所以它们不与“ URI” 识别。 它们只是数据值, 12. 3833。 我们在图表中也有属性、 属性或边缘 。 00:20:32.720 EN: These are also named also by URIs. Here also abbreviated has aria code. Longitude, latitude is mayor of. And also worth already noting is that they are directed. 中文初译: 这些也以“URIs”命名,这里缩略的有Aria代码。经度,纬度是市长。同样值得一提的是它们被指向了。 00:20:47.260 EN: It's perhaps not clearly visible in this slide. There's an arrow that goes from one node to the other node. And they have actually names, which we will see. 中文初译: 这张幻灯片中也许没有清晰可见。 有一支箭从一个节点飞到另一个节点, 而且它们实际上有名字, 我们会看到它。 00:20:58.900 EN: So important to remember, in RDF, these edges are always directed from one node. And there is this outgoing edge to another node, which has, therefore, an incoming edge. So this is a small graph. 中文初译: 要记住,在“RDF”中,这些边缘总是从一个节点向导。有一条向另一个节点的向外边缘,因此有一个即将到来的边。所以这是一个小图。 00:21:17.080 EN: And it simply says that Leipzig has a mayor, Burkhard Jung. And Burkhard Jung is a member of the Social Democratic Party. 中文初译: 简言之,莱比锡有一个市长,Burkhard Jung。而Burkhard Jung是社会民主党的党员。 00:21:26.820 EN: And Leipzig has an aria code and a longitude, latitude. Burkhard Jung was obviously born in some day. And Leipzig is located in Saxony. And Saxony is located in Germany. 中文初译: 莱比锡有亚里雅码和经度,纬度。布尔哈尔德·郑显然在某天出生。莱比锡位于萨克森。萨克森位于德国。 00:21:44.880 EN: So this is what this graph says. And it's a visual representation of this information using the RDF data model. 中文初译: 这就是这个图表所显示的。它用“RDF”数据模型来直观描述这些信息。 00:21:56.690 EN: So an important concept in RDF is the so-called RDF triple, sometimes also called statement. So that's a basic unit of information. 中文初译: 因此,在“RDF”中,一个重要的概念就是所谓的“RDF”三重声明,有时也称为声明。所以这是一个基本的信息单位。 00:22:08.570 EN: And it contains, essentially, two nodes and one edge. So remember, the edge is directed from the subject node to the object node. 中文初译: 它基本上包含两个节点和一个边缘。 所以请记住, 边缘是从主题节点到对象节点的方向。 00:22:22.090 EN: And the arrow or the edge is also called the predicate. And this is a little bit confusing. So we talk about nodes and edges when we speak about graphs. 中文初译: 箭头或边缘也被称为上游。这有点混乱。所以当我们谈论图表时,我们谈论节点和边缘。 00:22:35.470 EN: That's grounded in graph theory. And then we also talk about resources in RDF. So everything, actually, in RDF is a resource. 中文初译: 这是基于图表理论的。然后我们还在“RDF”中谈论资源问题。所以实际上,“RDF”中的一切都是一种资源。 00:22:45.350 EN: Then you have different types of resources. You have the identified resources. You have properties that are also subclasses of resources. 中文初译: 然后,你拥有不同的资源类型。你有已经确定的资源。你拥有同样属于亚类资源的属性。 00:22:57.330 EN: And you have literals. Here, this is not a literal, remember? So the green ones, oval, they are identified resource. 中文初译: 你们有各种经文,这里不是经文,难道你们记得吗?绿色的经文和古代的经文,是被鉴定的资源。 00:23:06.090 EN: And then we also speak about subject predicated object in the context of triples. And this is sort of important because, as you remember, the arrow is directed. 中文初译: 然后,我们又在三重角度上谈论 标定对象。这很重要,因为正如你所记得的, 箭是方向的。 00:23:19.410 EN: And so you need to distinguish the resource from which the arrow is departing and the resource to which the arrow is incoming. 中文初译: 所以你需要区分箭头离开的资源和箭头流入的资源。 00:23:30.330 EN: And these are separated by talking about subjects and objects. So the arrow always goes from the subject to the object. 中文初译: 这些是分开的,通过谈论对象和物体。所以箭头总是从对象到对象。 00:23:41.560 EN: And this terminology here, subject predicated object, is a little bit also grounded in linguistic categories. So when you have natural language sentences, you also talk about a subject. 中文初译: 这里的术语, 主题的标的对象, 有点也基于语言类别。 所以当您有自然的语言句子时, 您也会谈论一个主题 。 00:23:56.020 EN: You have a predicate and an object. So this structure comes, this triple structure comes, is also a basic unit in language. 中文初译: 你有一个前置物和一个对象。这个结构来了, 这个三重结构来了, 也是一个语言的基本单位。 00:24:06.600 EN: And yeah, it's reflected also in the fact that, in a way, RDF is a data model to encode statements, also natural language statements, where you have a subject about which something is said and the verb as a predicate. 中文初译: 是的,它也反映在一个事实中, 在某种程度上,“RDF”是一个数据模型 来编码语句, 也是自然语言语句, 在那里你有一个主题, 关于什么是说的东西,动词作为谓词。 00:24:29.660 EN: So here, we also consider this a subject. The subject resource can either be a URI, so it's always a resource. It can be an identified resource, so a URI-identified resource, or it can be a blank node. 中文初译: 因此,这里我们也考虑这个主题。 主题资源可以是“ URI ” , 所以它总是一种资源。 它可以是一个识别的资源, 所以是一个“ URI ” 的识别资源, 或者一个空白的节点 。 00:24:48.540 EN: The predicate is always a URI, so it's a URI-identified term. It's also called property, so just to confuse it a little bit more. 中文初译: 上游总是"URI",所以它是一个"URI"识别的术语。它也叫财产, 所以只是把它混淆一点。 00:25:02.180 EN: And then we have the object. And the object can, of course, be a resource. So a URI-identified resource, it can be a blank node, but in addition and in contrast to the subject resource, the object can also be a literal, as we saw already. 中文初译: 然后我们有了对象。当然,天体可以是一个资源。 所以一个“ URI” 标识的资源, 它可以是空白的节点, 但是与主题资源相比, 该对象也可以是字面的, 正如我们已经看到的那样。 00:25:20.960 EN: So we have a data value. And remember, from a literal, there is never an outgoing arrow and a predicate. So a literal is always sort of a leaf node, if you want, in the graph. 中文初译: 所以,我们有一个数据值。 并且记住, 从字面上看, 从来没有一个箭头和尾部。 因此在图表中, 字面上总是像叶节点一样, 如果你愿意的话。 00:25:41.610 EN: Yes, node and edges label should be unambiguous. So this is given by the property of URIs. Those are unambiguous identifiers. 中文初译: 是的,节点和边缘标签应该是明确的。这是由“URI”的属性给出的。这些是明确的标识符。 00:25:53.490 EN: But also the label, so as we see, so you can attach to each resource and predicate not just the URI, but also a label for it. 中文初译: 但标签,正如我们所看到的, 这样你就可以给每一种资源加上 并且不仅有“ URI ” 的标签, 而且还有它的标签。 00:26:06.230 EN: And it's, of course, useful if these labels are unambiguous and clear so that, yeah, so for people, but also for machines, they can query these properties, for instance, unambiguously. 中文初译: 当然,如果这些标签是明确和清晰的, 也是有益的。 因此,对人来说,对人来说,对机器来说, 他们可以对这些属性进行查询, 比如说,毫不含糊地。 00:26:23.470 EN: So what is a URI? And I'm sure that you have heard about this. A URI is the uniform, not to confuse with universal. 中文初译: 那么,什么是“URI”?我敢肯定你听说过这个。“URI”是制服,而不是与普遍性混淆。 00:26:32.050 EN: It's a uniform resource identifier, sometimes mistakenly called universal resource identifier. And they are used to create global unique names for resources in RDF. 中文初译: 它是一个统一的资源标识符, 有时被误称为通用资源标识符。 它们被用来在“ RDF” 中创建全球独特的资源名称 。 00:26:47.290 EN: So URIs is a technology that has existed for much longer than RDF. But the RDF data model builds on URIs or uses this identification scheme for the resources. 中文初译: 所以“URIs”是一种比“RDF”存在的时间长得多的技术。但“RDF”数据模型建立在“URI”的基础上,或者使用这种资源识别方法。 00:27:04.850 EN: Yes, so and every object, in a way, has therefore a clear identity, at least within the space of the web. And they are globally unique or globally unambiguous. 中文初译: 因此,从某种意义上说,每个物体都有清晰的特性,至少在网络空间之内。 它们在全球是独一无二或毫不含糊的。 00:27:23.470 EN: If you use an identifier, HTTP example.org slash book one, and someone else uses the same identifier, there is sort of an understanding that these are actually the same things, although logically you can debate whether that's true or not. 中文初译: 如果您使用识别符号, HTTP 例如. org slash book one, 而其他人也使用相同的识别符号, 有一种理解是, 虽然逻辑上你可以辩论这是否属实。 00:27:43.390 EN: But in RDF, the RDF data model would simply merge then everything that is stated about the example.org book one in one data set with whatever is stated about example.org book one in the other data set. 中文初译: 在“RDF”中,“RDF”数据模型将简单合并所有关于举例.org 在一个数据集中一书的描述,而任何关于例如.org 另一数据集中一书的描述。 00:28:04.380 EN: And of course, as you know from the books domain, we have also other identifiers. So URIs are just one identification scheme. 中文初译: 当然,正如你从书域中知道的,我们也有其他识别资料。所以“URIs”只是一种识别方法。 00:28:15.900 EN: The International Standard Book Name, I think it's called, ISBN, is yet another identification scheme among many, many others. 中文初译: 国际标准书名,我想它叫ISBN, 是其它许多其他的又一个识别计划。 00:28:24.730 EN: But in contrast to ISBNs and other identification schemes, URIs have an advantage because they are also, at least they have the potential to be globally or resolvable on the web. 中文初译: 但与ISBNs和其他识别计划相比,“URIs”也有优势,因为它们也是,至少它们具有在全球范围或可在网上解决的潜力。 00:28:44.710 EN: So as you know, if you type something like example.org book one, the web provides a mechanism to resolve this URI on the web and therefore obtain more information without needing a third-party lookup system, like in the case of ISBNs. 中文初译: 所以,你知道,如果你输入像 example.org book one这样的字, 网络提供了一个机制来解决这个网络上的“URI”问题, 00:29:07.850 EN: You would have to have a service where you give the ISBN, which is just a number. Well, not actually a number, but numbers and I think dashes in between. 中文初译: 你必须有一个服务 在那里你给ISBN,这只是一个数字。嗯,其实不是一个数字,但数字和我想数字会冲进中间。 00:29:18.010 EN: So you can look these things up, of course, so to get more information about the ISBN, so what book it identifies. 中文初译: 所以,你可以研究这些东西,当然, 以便获得更多关于 ISBN的信息, 那么它识别的书是什么。 00:29:24.990 EN: But you need a service that implements that lookup mechanism versus URIs just leverage the web itself and the internet. 中文初译: 但是,你需要一个服务 来实施这个搜索机制 而不是“URIs” 只是利用网络本身和互联网。 00:29:36.610 EN: So you can just paste a URI in the browser and assuming it locate, there is a location for that URI on the web. 中文初译: 你可以在浏览器上粘贴一个"URI" 假设它定位了, 在网上有一个“URI”的位置。 00:29:48.790 EN: So there is either a URL it translates to or the URI is actually already directly a URL. Then you will have just leveraging the web and internet technologies. 中文初译: 所以要么是它翻译到的 URL,要么就是“URI”实际上已经直接是一个 URL。然后你就可以利用网络和互联网技术了。 00:30:07.390 EN: You will find the location where you can find more information about this resource on the web. So that's quite powerful because you don't need a third party system for a lookup mechanism for these identifiers. 中文初译: 您可以在网络上找到更多关于此资源的信息。 因此这个功能相当强大, 因为不需要第三方系统来查找这些识别资料。 00:30:28.940 EN: So the URI syntax, again, I think this should be clear. It's an extension of the URL concept. So not every URI actually denotes a web document. 中文初译: 所以“ URI” 语法, 我认为这应该很清楚。 这是 URL 概念的延伸 。 所以不是每个“ URI” 都表示网络文件 。 00:30:41.840 EN: So it's not necessarily so that you can resolve a URI on the web. So for this, you need a location, a URL, a uniform resource location. 中文初译: 所以这并不一定是为了解决网络上的“URI ” 。 因此,你需要一个位置,一个 URL, 一个统一的资源位置。 00:30:55.820 EN: These are the actual web addresses, like here slidewiki.org presentation something, or I don't know www.google.org. 中文初译: 这些是实际的网址,例如这里的幻灯片wiki.org演示文稿,或者我不知道www.google.org。 00:31:06.720 EN: These are URLs and they have, if you type them in the browser, you get either an error or on the actual website. 中文初译: 这些是 URL 。 如果您在浏览器中输入, 它们就会在实际网站上发现错误 。 00:31:15.300 EN: And with URIs, that's not the case. If it could be that it resolves to a location and therefore a website potentially an error, but it can also simply be an identifier without a location. 中文初译: 使用“URIs ” , 情况并非如此。 如果它能解决到某个位置, 因此网站可能会出错, 但它也可以是一个没有位置的识别符号。 00:31:33.800 EN: Usually they have the schema that starts with HTTP. So as a schema URI in the RDF data model, they can use all of this. 中文初译: 通常他们有HTTP开头的图案。因此,作为“RDF”数据模型中的图案“URI”,他们可以使用所有这些方法。 00:31:47.120 EN: And in fact, also mail too is sometimes used to encode the email addresses of people in RDF. Also file could be used if you want to directly access a file using the file protocol. 中文初译: 事实上,邮件有时也被用来在“RDF”中编码人们的电子邮件地址。如果您想要使用文件协议直接访问文件,也可以使用文件。 00:32:02.340 EN: But most URIs in the RDF data model will start with the schema HTTP. And then as you know, you have this typical hierarchical structure, the scheme, and then some authority that will be the domain name. 中文初译: 但“RDF”数据模型中的大多数“URIs”将首先使用Schema HTTP。然后,如你所知,你有一个典型的等级结构,这个计划, 然后是某种权威,它将成为域名。 00:32:18.100 EN: So google.org and then a path, a file path. And then in the end, some queries potentially, and also a fragment. 中文初译: org,然后是一条路径,一个文件路径。最后,一些可能查询,还有一个碎片。 00:32:27.480 EN: To this, I like to say that there was a long discussion in the community about the so-called cool URIs. And generally it's so that queries shouldn't be included in cool URIs. 中文初译: 对于这一点,我想说社区中长期讨论所谓的“URIs ” 。 一般来说,这样就不应该把查询包括在“URIs”中了。 00:32:44.440 EN: That has the reason for this is that URIs should also be stable. They shouldn't change too widely and too often. 中文初译: 之所以如此,是因为“URIs”也应该稳定。它们不应该变化太广泛和频繁。 00:32:52.960 EN: Ideally it wouldn't change at all because once people start to use URIs, your URIs to state additional things about resources or for instance, query for URIs in other systems, then if you change them, then things will break. 中文初译: 最好不会改变 因为一旦人们开始使用“URIs”, 你的“URIs”来说明额外的资源信息, 或者比如在其他系统中查询“URIs”, 00:33:11.940 EN: So they should have, as identifier systems should provide, they should provide some stability. And query parameters are too close to the implementation to the system as is for instance, information in the path about HTML websites. 中文初译: 因此,它们应该像标识系统应该提供的那样,具有一定的稳定性。 查询参数与该系统的实施太接近,例如HTML网站路径中的信息。 00:33:33.360 EN: So that also, or PHP. Sometimes you have .html, .php or the .jsp. And this is all too technology specific. And since technology of system might change, so in the 10 years from now, you might move from PHP to, I don't know, something else. 中文初译: 也就是说PHP。有时你有.html、.php或.jsp。这太具技术特质了。由于系统技术可能会改变,所以在10年后的十年里,你可能会从PHP转到(我不知道)其他的东西。 00:33:55.630 EN: And then this path would have to be changed as well. And that would break your URIs. So you have to be careful about when you construct URIs to also try to be the least semantic as possible in these URIs. 中文初译: 然后这条路也必须改变。那会打破你的“URIs ” 。 所以你必须小心 当你构建“URIs ” 的时候, 你也要尽量成为这些“URIs”中 尽可能少的语义。 00:34:15.980 EN: So not put too much meaning and certainly not include any aspects of technologies. Also the path. So, slash, slash, slash, this can change if you move your, I mean, worst case, if you move your files on the system, of course it changed, but also in REST APIs, if you change slightly your data model, it could also have implications on your URI syntax. 中文初译: 所以不要说太多意义,当然也不要包括任何技术方面。还有路径。因此,斜线,斜线,斜线,斜线,如果移动最坏的情况,如果移动系统上的文件,当然会改变,也可以改变,但是在REST APIs中,如果你稍微修改数据模型,也会对“URI”语法产生影响。 00:34:48.220 EN: So these are just a few things to remember in URI construction, in particular of course, if you use these technologies in productive systems that are supposed to run for a long time, then these things matter. 中文初译: 所以,这些只是“URI”建设中需要记住的几件事情, 特别是,如果你在生产系统中使用这些技术, 这些技术应该运行很长时间, 那么这些都很重要。 00:35:03.560 EN: If you just build a prototype or you do an exercise in class, then of course this is a little bit theoretical. And yeah, sometimes you need to self-define your URI. 中文初译: 如果你只是做一个原型或者在课堂上做个练习, 当然这当然有点理论性。是的,有时你需要自我防守你的“URI”。 00:35:17.980 EN: So sometimes things are already identified. So think about Joe Biden to names a person has, most likely I didn't check, but I am pretty sure that Joe Biden has a Wikidata URI and or also Wikipedia URI. 中文初译: 所以有时事情已经被确定。所以想想乔・拜登的名字, 很可能我没有检查, 但我很肯定乔・拜登有维基数据“团结”或者维基百科“团结”。 00:35:38.980 EN: You could, for instance, if you want to say something about Joe Biden or you want to query Wikidata for Joe Biden to get some information that Wikidata has about Joe Biden, then you would use this URI directly. 中文初译: 比如说,如果你想对乔・拜登说点什么 或者你想向维基数据查询 找乔・拜登了解维基数据 维基数据关于乔・拜登的信息 00:35:53.600 EN: So Joe Biden has a URI, fine. But some things like maybe yourself or some concept that is not widely known and so, or a concept that you have created, they don't have a URI yet. 中文初译: Joe Biden有一个“URI ” ( URI),很好。但有些东西,比如你自己或者一些尚未广为人知的概念, 或者你所创造的概念, 还没有“URI ” ( URI)的概念。 00:36:08.120 EN: So in this case is you have to create your own URI and this is a pretty free model. So usually of course it's best if you use your own namespace or a website or a domain with a namespace that you sort of control. 中文初译: 所以在这种情况下,您必须创建自己的“ URI ” 并且这是一个相当自由的模式。 因此通常最好使用您自己的命名空间或网站, 或者有您可以控制的名称空间的域名 。 00:36:26.780 EN: So that could be in my case, markerstalker.com or something else, but not google.org or google.com because that's not in your own control. 中文初译: 因此,这可能是在我的情况下, 标记跟踪器.com或其他的东西, 但不是google.org或google.com, 因为这不是你自己控制的。 00:36:40.600 EN: And this way you can, I wouldn't say you can avoid naming collision because of course anyone can use any domain that it's not like with URLs where you actually buy the domain and that's in your control and only you can create path is under this domain and everything else that someone else would create wouldn't resolve. 中文初译: 这样你就可以避免命名相撞, 因为当然任何人都可以使用任何域名, 因为它与URL不同, 而URL其实是买下域名的, 这是由你控制的, 只有你可以创造路径, 并且所有其他别人创造的东西 都无法解决。 00:37:04.140 EN: In your eyes, anyone can use these authorities. So, and there is no restriction on the, so here. So someone else could use markerstalker.com and have naming collisions. 中文初译: 在你眼里,任何人都可以使用这些权威。所以,对于这些权力没有限制,所以这里是这样。这样其他人就可以使用标记跟踪器.com, 并命名碰撞。 00:37:20.820 EN: But if everybody uses this best practice to try to have namespaces for your own vocabularies, for your own terms, for your own resources, within the namespace that you control, this is a way to sort of reduce anyway the chances for naming collisions. 中文初译: 但如果每个人都使用这个最佳做法 尝试为自己的词汇, 以你自己的条件, 以你自己的资源, 在你控制的命名空间内, 拥有命名空间, 这是一种方法来减少 碰撞的命名机会。 00:37:45.980 EN: And naming collisions are essentially if you use the same resource, the same identifier for different resources. 中文初译: 命名碰撞基本上是 如果你使用同样的资源, 不同的资源使用相同的识别符。 00:37:54.500 EN: If you use the same identifier for the same resource, that's great and that's in fact what you should do, but you should avoid that two different, that actually different things have the same identifier. 中文初译: 如果你对同一资源使用相同的识别码,那很好,事实上你应该这样做, 但你应该避免两种不同的识别码, 实际上不同的东西有相同的识别码。 00:38:11.820 EN: Yeah, and what also the slide highlights here is that you should separate the URI, so the identifier for the actual real world thing for the resource itself. 中文初译: 是的,幻灯片也突出显示的是 你应该把“URI”分离出来, 所以对于资源本身来说, 真实世界的标识符。 00:38:23.560 EN: For instance, the person and its documentation. So usually then a URI should, best practice is that a URI also has some location on the web so that you can look this URI up or you can get more information about the identified real world thing. 中文初译: 例如,个人及其文件。通常“URI”应该有一个“URI ” , 最佳做法是“URI”在网络上也有一定的位置, 这样您就可以查看这个“URI ” , 或者获得更多关于被识别的真实世界的信息。 00:38:44.640 EN: And that is for humans anyway, generally an HTML page where you can find more information about the resource. And these things, they should be different. 中文初译: 也就是说,对于人类来说,通常是一个 HTML 页面, 你可以找到更多关于资源的信息。 而这些东西应该不同。 00:38:54.800 EN: So the URI of the thing should be different from the URL of the HTML page. And there are various ways how to introduce these differences. 中文初译: 所以“ URI” 与 HTML 页面的 URL 应该是不同的。 而且有多种方法可以引入这些差异 。 00:39:06.800 EN: To separate, to have clearly separate strings. And here is an example below, Shakespeare Othello. So here the first one is essentially the Wikipedia page of Othello. 中文初译: 要分开, 需要明确分开的字符串。 下面请举个例子, 莎士比亚奥瑟罗。 所以这里的第一张基本上是 Othello 维基百科页面。 00:39:23.360 EN: And it's bad to use this one as a URI. And there is a question again here, why? Why is it bad to use httpd.vikipedia.org. 中文初译: 使用这个作为“URI”是件坏事。这里又有一个问题,为什么?使用httpd.vikipedia.org为什么不好呢? 00:39:36.680 EN: Viki Othello as a URI for Shakespeare's Othello. Maybe I'll leave again a minute here and let's see if you get some answers. 中文初译: 维基Othello是莎士比亚奥赛罗的"团结" 也许我会再离开一分钟 00:39:52.440 EN: Oh, I see Yaser, yes. Is this Joe Biden? Yes, exactly, good example. So here we have Joe Biden on the, so Joe Biden has a URI. 中文初译: 哦,我看到了亚瑟,是的。这是乔・拜登吗?正是,很好的例子。我们这里有乔・拜登,所以乔・拜登有一个“团结” 00:40:10.460 EN: And missing attached fragment. Yeah, that's, I guess, what the slide suggests. Now I, using the, I'm here, please start from scratch. Okay, that wasn't smart. 中文初译: 是的,我猜这就是幻灯片暗示的。现在我用这个,我在这里,请从头开始。好吧,那不聪明。 00:40:38.880 EN: We were here. So Joe Biden, and you can see that Yaser in the chat has copied, what I would say is the URL of Joe Biden. 中文初译: 我们来过这里。所以Joe Biden,你可以看到在聊天中的Yaser复制了, 我想说的是Joe Biden的网址。 00:40:51.940 EN: So let me do this this time. We can also demonstrate this issue here on exactly this page. So as you can see, this is a Viki Q6279 here. 中文初译: 因此,让我这次这样做。我们也可以在这个页面上展示这个问题。你们可以看到,这里的维基Q6279。 00:41:08.400 EN: And this is the URL of the website describing Joe Biden with HTML. So for humans, but actually Joe Biden has a so-called concept URI. 中文初译: 这是介绍Joe Biden使用HTML的网址的 URL。对于人类来说, 但实际上Joe Biden有一个所谓的“URI”概念。 00:41:23.640 EN: And that's the URI of Joe Biden. So if you use, if you use this one, that's the correct URI identifying Joe Biden as a person. 中文初译: 这就是Joe Biden的"URI"。所以如果你使用,如果你使用这个,那就是正确的"URI" 识别Joe Biden的人格。 00:41:42.960 EN: So Joe Biden as the real world thing, the resource. So see the differences here that the Viki data uses entity slash entity slash Q, that's the identifier, versus the website is at slash Viki slash and then the same identifier. 中文初译: Joe Biden是真实世界的事物,资源。请看维基数据使用实体刀切实体刀切Q(即识别符)的不同之处, 而网站是刀切维基刀切,然后是相同的标识符。 00:42:04.100 EN: So that's the difference. This is the URL of the identifying the sort of the location or showing the location, pointing at the location on the web of this website. 中文初译: 这就是区别。这是用来识别位置或显示位置的 URL, 指向这个网站的网页上的位置。 00:42:15.040 EN: So this is a documentation of Biden versus this one is the URI identifying the real world thing that's the person Joe Biden. 中文初译: 这是一本关于拜登和这个文件的文献, 上面写着“URI” 识别真实世界的东西, 就是Joe Biden。 00:42:28.320 EN: And if you resolve this entity Q6279, I'm sure that you end up on the Viki page. So because I am a person and I did this with a client, a browser, the system actually resolves this URI then also on the documentation that is designed for people. 中文初译: 如果你解决这个实体 Q6279, 我肯定你最终会出现在维基页面上。所以因为我是一个人,我用一个客户,一个浏览器, 这个系统实际上解决了这个“URI”, 然后也解决了为人们设计的文档。 00:42:51.880 EN: But as you probably know, with content negotiation, you can also retrieve when you resolve a URI, you can also obtain the same description here, the same information, but in a more machine readable format, could be RAF, could be a JSON or JSON-LD and so on. 中文初译: 但正如你可能知道的,通过内容谈判, 当你解决了“URI”, 你也可以检索到相同的描述, 同样的信息, 但是以更机器可读的格式, 可能是RAF, 可以是JSON, 或者“JSON-LD”等等。 00:43:16.300 EN: So this is the reason and essentially this is a demonstration of the same thing here. Viki, Otello, I'm not sure about Wikipedia, but I think they also have concept URIs, which are not slash Viki slash Otello, but I'm not sure if they use also slash entities, slash Otello or something else. 中文初译: 因此,这就是原因,而且本质上,这是同样的事情的证明。维基百科,Otello,我不确定维基百科, 但我也认为他们也有“URI”的概念, 而不是“URI”概念, 它不是“Slash Viki slish Otello”, 但我不确定它们是否还使用“Slish Viki scrap Otello”或其他东西。 00:43:36.560 EN: That will be one way to do it, as we just saw in Wikidata. Another way how it's done here, I don't particularly agree that's quite elegant. 中文初译: 这将会是做到这一点的一种方法,正如我们在维基数据中看到的。另外一种方式,我并不特别同意这是相当优雅的。 00:43:47.680 EN: I think the way Wikidata did it with slash entities, slash Otello and slash Viki and slash Otello is nice or better practice than just appending a fragment URI, but they are different strings and the point is that these strings, the one identifying the real world thing and the one locating the description HTML page on the web, they are in fact different. 中文初译: 我认为维基数据与砍刀实体、砍刀Otello、砍刀Viki和砍刀Otello的操作方式, 00:44:18.740 EN: So that's the main point. But again, there was a long discussion about cool URIs, so you can dig that up and you will find fun discussions on what exactly is a cool URI. 中文初译: 这就是要点。但是,关于“URIs”的争论是很长的, 所以你可以把它找出来, 你会发现关于什么是“URI”的争论很有趣。 00:44:34.500 EN: So we had URIs and now we also have something like IRIs. This is the Internationalized Resource Identifier and it's essentially a generalization of the URI concept. 中文初译: 所以,我们有了“URIs”,现在我们也有类似“IRI”的东西。这是国际化的资源识别符, 基本上就是“URI”概念的概括。 00:44:48.820 EN: And the biggest difference is then in IRIs, you can use Unicode characters, as you can see here, slash umlaut, that's not an ASCII character, which is the set of character used in URIs. 中文初译: 最大的区别在于“IRIs”, 您可以使用 Unicode 字符, 正如您在这里看到的, slash umlaut, 这不是 ASCII 字符, 这是“ URIS” 中使用的字符组 。 00:45:05.800 EN: You can also use, I don't know what language that is, but any other language, I think it covers, Unicode covers pretty much any other character known to mankind. 中文初译: 你也可以使用,我不知道那是什么语言, 但是任何其他语言,我想它都涵盖了, Unicode 几乎覆盖了人类所知的其他任何特征。 00:45:17.300 EN: So that's a nice way, of course, if you want to encode this, not just URIs, but also text, of course, as you know from Unicode. 中文初译: 所以,这是一个很好的方法,当然, 如果你想编码这个, 不仅仅是“URIS”, 而且还有文本, 当然,正如你从Unicode中知道的。 00:45:34.820 EN: So then we talked about literals. So we had URIs as a identification scheme in the RDF model and why this is important and how it's used, how it's constructed and some best practices there that we discussed. 中文初译: 然后我们谈论了字面意思。所以我们在“RDF”模型中用“URIs”作为识别方法, 为什么这很重要, 以及它是如何使用的, 它是如何构建的, 以及我们讨论过的一些最佳做法。 00:45:51.480 EN: So the other important element of the RDF data model is literals. And as I already mentioned, literals are used to encode data values. 中文初译: 因此“ RDF” 数据模型的另外一个重要元素是字面的。 正如我已经提到的, 字面用来编码数据值。 00:46:03.120 EN: They are represented in RDF model as strings. So they are actually, RDF data is textual data. So it's essentially text document. 中文初译: 它们以字符串表示在“RDF”模型中。所以它们实际上是,“RDF”数据是文字数据。因此它基本上是文本文档。 00:46:15.240 EN: And so all values, all data values of any type are encoded as strings. And then you have an interpretation of this string to the correct data type, which is given by, you know, explicitly given by some data type. 中文初译: 所以所有值, 任何类型的所有数据值都编码为字符串。 然后您可以将此字符串解释为正确的数据类型, 由某种数据类型明确给出。 00:46:36.780 EN: And I will have more to say about this later. And if data type is not explicitly given, then these strings are in fact, treated as strings. 中文初译: 稍后我将对此有更多的话要说。如果数据类型没有明确给出,那么这些字符串实际上是作为字符串处理的。 00:46:49.360 EN: And what I also already mentioned is that literals can never be an origin of a node, of an origin of a node, the origin of an edge, I'm not sure. 中文初译: 我也已经提到过 字面上不可能是节点的起源, 节点的起源,边缘的起源, 我不确定。 00:47:01.120 EN: So the point is literals, there can never be an edge, so a property or attribute departing from a literal. So it can never be a subject node, so to say, in an RDF graph. 中文初译: 意思是字面意思, 永远不可能有边, 所以属性或属性从字面意思出发。 因此在“ RDF” 图中, 它永远不会是一个主题节点。 00:47:19.700 EN: And also edges cannot be labeled with literals. Yes, so edges are always labeled, they're always identified by URIs. 中文初译: 边缘也不能用字面标注。 是的, 所以边缘总是贴上标签, 它们总是被“ URIs” 识别 。 00:47:31.200 EN: So here an example, we had Leipzig, and the mayor of Leipzig, Burkhard Jung, but then we have here literals, remember these red rectangular boxes here. 中文初译: 举个例子,我们有莱比锡 莱比锡的市长,布哈尔德·郑, 但是我们在这里有了字面文章, 记得这里的红色矩形盒子。 00:47:45.220 EN: And you can see now here, data type. So these are now explicitly typed literals. So as I said in the slide, so if you have the data values are represented as strings in RDF, so 12.3833 would be a string, but the interpretation of the string is a double 中文初译: 现在您可以看到这里的数据类型。 因此这些是明确的输入字的文字。 正如我在幻灯片中所说, 所以如果数据值在“ RDF” 中被以字符串表示, 那么12.3833 将是一个字符串, 但是对字符串的解释是双 00:48:14.360 EN: because there is an explicit data type stated here. So that systems need to interpret this string as a double, and that has of course implications. 中文初译: 因为这里有明确的数据类型。 因此系统需要将这个字符串解释为双倍, 这当然会产生影响 。 00:48:30.120 EN: So maybe here with double, it's not pretty clear, but certainly here for data, so that's a string, numbers and dashes. 中文初译: 所以,也许这里的双倍, 并不十分清楚, 但肯定在这里的数据, 所以这是一个字符串,数字和破折号。 00:48:43.600 EN: But if you say that this is actually, this string should be interpreted as a date, this has huge implication on systems, for instance, sorting these literals, or also interpreting these literals that this is the year, and this is the month, and this is the day according to the ISO standard for dates. 中文初译: 但如果你说这是真的, 这个字符串应该被解释为一个日期, 这对系统有巨大的影响, 比如分类这些字眼, 或者解释这些字眼, 也就是今年是这个月份, 这是按照ISO 标准计算日期的一天。 00:49:06.320 EN: So yeah, it matters whether you explicitly type these strings as date or a string, or as double would make any sense. 中文初译: 所以是,你把这些字符串明确输入日期还是字符串, 还是双倍键入, 是否合理都很重要。 00:49:19.760 EN: Good, yes. Can I ask a question here? Sure. So the object and the subject will be URIs usually. A means, and the object and the subject can have then literals. 中文初译: 很好,是的。我可以在这里问一个问题吗?当然。所以对象和主题通常都是“URIs ” 。一种手段,这个对象和主题可以有文字。 00:49:40.880 EN: Okay, so the subject is always a URI. So the subject here are the green nodes. The object can be a URI, I wouldn't say that it's mostly a URI, I don't have any statistics in idea of data sets, but the object can be a URI. 中文初译: 主题总是“URI”。所以这里的主题就是绿色节点。对象可以是“URI”,我不会说它大部分是“URI”,我对数据集没有统计,但对象可以是“URI”。 00:50:02.740 EN: In this case, you have here, Burkhart Jung. Actually, this is an interesting case. So you have here two relations, is Mayor off and has Mayor. 中文初译: 事实上,这是一个有趣的案例。所以你有两个关系, 市长不在了,市长也在这里。 00:50:16.800 EN: One time it's pointing from this node to this node. I hope you see my cursor. Yeah, okay. Yeah, it's directed from this node to this node. 中文初译: 有一次它把节点指向这个节点 我希望你看到我的光标 00:50:27.960 EN: And the other time it's the other way around. And you can see that has Mayor and is Mayor off are inverse properties. 中文初译: 换句话说 你可以看到市长和市长是相反的属性 00:50:38.080 EN: And so even though, so Leipzig is a resource, Burkhart Jung is a resource, right? But in one triple Leipzig is a subject and the other triple Leipzig is an object. 中文初译: 因此,尽管莱比锡是一个资源,但郑布尔哈特是一个资源,对吗? 但是在一个3倍的雷比锡是主题,另一个3倍的莱比锡是目标。 00:50:52.120 EN: So remember that triples is always a unit that connects through an edge to nodes. So in this, if you discard now the literals here, if you look only at this part, these are actually two statements, two triples. 中文初译: 记住,三重总是一个连接到节点边缘的单位。 因此,在这里,如果你现在丢弃这里的字面文字, 如果你只看这部分的话, 这些实际上是两个对白, 两个对白。 00:51:10.140 EN: So one time Leipzig is a subject and Burkhart Jung is the object related by the has Mayor property. And in the other statement or triple, Burkhart Jung is a subject and Leipzig is the object, and they are related by the is Mayor property relating the subject to the object. 中文初译: 因此,有一次莱比锡是一个对象,布尔卡特·郑是市长拥有财产的对象。在另一则声明或三重声明中,布尔卡特·郑是目标主体,莱比锡是目标主体,他们与目标相关联的是市长财产。 00:51:34.880 EN: So it's interesting because, depending on which property you're looking at here, subject, so Leipzig can be the subject or Leipzig can be the object. 中文初译: 所以很有趣,因为,取决于你在这里看的 财产,主题, 所以莱比锡可以是主体 或者莱比锡可以成为对象。 00:51:47.180 EN: The object can be a URI, which is the case here. We had a Leipzig as an object, a Leipzig as a subject. So that's possible, but the object can also be a data, sorry, a data type, a literal, which is the case in these three red boxes. 中文初译: 对象可以是“ URI ” , 这里就是这种情况。 我们有一个 Leipzig 作为对象, 一个leipzig 作为主体。 所以这是可能的, 但是对象也可以是一个数据, 抱歉, 一个数据类型, 一个字面的, 这三件红色框中都是这样 。 00:52:06.520 EN: Sorry about the noise, they are drilling upstairs. So this is the case here. So these are literals. And so the object can also be a literal. 中文初译: 抱歉噪音,他们正在楼上钻井。这里的情况是这样。所以这些是字面的。这个对象也可以是字面的。 00:52:22.580 EN: Was that clear? Thank you. Okay, good. Any other question? How are we doing in time? Good, I think we have half an hour. 中文初译: 清楚了吗?谢谢。好的,还有什么问题吗?我们如何及时进行?很好,我想我们还有半小时时间。 00:52:45.630 EN: So maybe we should speed up a little bit. Good, then this slide introduces already a little bit turtle syntax because we need it later on in this chapter in section two of these lectures. 中文初译: 所以也许我们应该加快一点。好的,那么这个幻灯片已经引入了一点点海龟语法, 因为我们在本章第二部分的演讲中需要它。 00:53:02.110 EN: The next chapter, section three, I think it's three, yes, we'll talk even more about syntaxes, but just for you to a little bit understand the next slides. 中文初译: 下一章,第三节,我想是第三章, 是的,我们会讨论更多关于语法的话题, 但只是让你稍微理解一下接下来的幻灯片。 00:53:13.210 EN: So I'm introducing here the turtle syntax. So we have seen the RDF data and some examples visually represented as graphs. 中文初译: 因此,我在这里介绍龟语法。所以我们已经看到“RDF”数据, 以及一些以图示显示的例子。 00:53:23.490 EN: So if you remember the Leipzig example. This is of course not good for machines to process. You don't want to feed such graphical representations of RDF data into a system that would be terrible to process. 中文初译: 所以,如果你还记得Leipzig 的例子。 这当然不利于机器处理。 您不想将“ RDF” 数据的图形表达方式 输入到一个对处理来说非常可怕的系统中。 00:53:40.850 EN: So what we need to do is we need to be able to serialize RDF data. So data structured using the RDF data model into strings. 中文初译: 所以我们需要做的是,我们需要能够将“RDF”数据序列化。所以数据结构是用“RDF”数据模型构建成字符串的。 00:53:51.950 EN: Remember RDF data is essentially textual data. So it's an RDF document is a text document. So what we do here with serialization syntaxes or languages is to transform an RDF set of triples into a string. 中文初译: 记住“ RDF” 数据本质上是文字数据。 所以这是一个“ RDF” 文档是一个文本文档。 因此我们在这里使用序列语法或语言将三重字符串转换为字符串 。 00:54:16.930 EN: And there are various techniques or approaches to do this. And turtle is one of these languages to serialize RDF triples into strings. 中文初译: 这样做有多种技术或方法。海龟是将这些“RDF”三联成字符串的语言之一。 00:54:29.630 EN: Turtle is the terse, the RDF triple language. It has a number of language features which can run through quickly. 中文初译: “Turtle”是“RDF”的三重语言。 它有许多语言特征可以快速穿透。 00:54:40.150 EN: So URIs are encoded in angle brackets. You can see here, this is a URI with this angle brackets open and closing as you know from XML. The literals in quotes. 中文初译: 所以“ URIs” 是用括号编码的。 您可以在这里看到, 这是一个“ URI ” , 按您在“ XML ” 中所知道的那样, 括号打开并关闭。 引文中的字句 。 00:54:55.390 EN: Shown here, this is a language tag which we will talk a little bit more later. And it is already, we saw this is the way how you can encode the data type of a literal. 中文初译: 在这里显示, 这是一个语言标记, 我们稍后再谈。 我们已经看到这是一种方式, 您可以用它来编码一个字面的数据类型 。 00:55:09.790 EN: And then we have also this feature that statements are terminated by a dot. You can see here very small, perhaps the dot at the end of the statement. 中文初译: 然后我们也有这样一个特征,即声明被一个点终止。你可以看到这里非常小,或许是发言结尾的点。 00:55:21.870 EN: So here you have the subject, URI, the predicates URI encoded with angle brackets because they are URIs. Then you have the object. This is a literal quoted. 中文初译: 这里的主题是“ URI ” , 上游的“ URI” 编码为括号, 因为它们是“ URI ” 。 然后你有对象。 这是文字引用 。 00:55:37.030 EN: And the language tag here at the end DE, meaning that this literal is in German language. And then the dot at the end that terminates this triple in turtle syntax. 中文初译: 语言标记在底部 DE,意思是这个字是德文。最后的点 结束了海龟语法中的三倍。 00:55:54.550 EN: Yes, white space and line breaks are ignored outside identifiers. And this also is a W3C recommendation. You can look it up on the web. 中文初译: 是的, 白色空格和换行符会被忽略外部标识符。 这也是一个 W3C 建议。 您可以在网络上查看它 。 00:56:05.430 EN: A few other features of this language, you can abbreviate it. So you can introduce prefixes and then abbreviations for this URIs and namespaces. 中文初译: 此语言的其他几个特性, 您可以缩略下来。 这样您就可以为此“ URIs” 和命名空间引入前缀和缩写 。 00:56:20.300 EN: So for instance, we have here a DBPDR resource. We introduce a prefix, DBR colon and then sort of the namespace for these resources. 中文初译: 例如,我们这里有一个 DBPDR 资源。 我们引入了一个前缀, DBR 结肠, 然后对这些资源的命名空间排序 。 00:56:36.850 EN: So that will be DBPDR.org slash resource slash. And usually at the end there is something more, but this is the namespace for it. 中文初译: 这将是 DBPDR. org 的斜线资源斜线。 通常最后会有更多的东西, 但这是它的命名空间 。 00:56:46.070 EN: And then you can transform something cryptic like this into something more elegant and short like this. So you can see that essentially what we do is we drop this namespace DBPDR resource. 中文初译: 然后你就可以把这种神秘的东西 变成更优雅和短短的。所以你可以看到,我们所做的基本上就是 丢弃这个命名空间 DBPDR 资源。 00:57:02.850 EN: We put it, we introduce a prefix for it, DBR. And then we use the prefix here instead of writing out the whole URI. 中文初译: 然后我们在这里用前缀 而不是写出整个"URI" 00:57:15.770 EN: So we do this both for the DBI resource libc, DBR resource libc and the RDFS label. So this label is a property in the RDFS schema. 中文初译: 因此,我们这样做既是为了DBI资源解放,也是为了DBR资源解放和“RDFS”标签。所以这个标签是“RDFS”模型中的一种属性。 00:57:33.110 EN: And we will tell more about RDFS in the future lecture. So these are essentially equivalent triples, one abbreviated and the other not. 中文初译: 我们将在未来的演讲中更多地讲解“RDFS ” ( RDFS ) 。 因此,这些基本上是等效的三倍,一个是缩写,另一个不是。 00:57:46.150 EN: So you can do even more abbreviation and this is all sort of syntactic sugar. So this is all to make it more readable, more compact, more readable for humans. 中文初译: 所以,你可以做更多的缩写, 这都是一种合成糖。所以这一切都是为了让它更易读, 更紧凑, 对人类更易读。 00:57:57.710 EN: I would say for machines there's absolutely no difference. You can translate one into the other very easily, but for humans it's nicer if you just have more compact form. 中文初译: 我想说机器是完全没有区别的。你可以很容易地把一个转换成另一个, 但是对于人类来说,如果你有更紧凑的形式,它会更好。 00:58:10.050 EN: So here is yet another one. So when you state different things about same subject, you can also drop the subjects. It's only written on the first line, subsequent lines, don't mention lines anymore. 中文初译: 这是另一个问题。所以当你对同一主题说不同的话时, 你也可以放弃主题。它只写在第一行, 其后的线上, 不再提到线条。 00:58:27.290 EN: But you need to introduce this semicolon at the end because this is the terminal character for statements or triples in the case the subject is mentioned only once in this set of triples. 中文初译: 但是您需要在结尾处引入此分号, 因为这是语句的终端字符, 或者在主题只被提及过一次的情况下, 在一组三重中, 这是三重中的结束字符 。 00:58:49.030 EN: So you can see here, every time you want to reuse sort of the same subject, you terminate the first line, the previous line with semicolon at the end, you put again a period. 中文初译: 所以,你可以看到这里,每次你想重复使用相同主题的排序时, 你就会终止第一行, 上一条在结尾处加上分号的直线, 然后再加一个句子。 00:59:01.890 EN: So in this case, this is an abbreviation of the subject sort of, and you can do this even more if even further, if the subject and the predicate are same in two different triples, then you can list the objects by separating them with a comma. 中文初译: 因此,在这个案例中,这是一个主题的缩略语, 你可以做得更多, 如果更进一步的话, 如果主体和上游在两个不同的三重中是相同的, 那么您就可以通过用逗号将其分隔来列出对象 。 00:59:27.070 EN: So these are essentially two triples. So one is this one, Leipzig is located in Saxony, and the other one is Leipzig is located in Germany. 中文初译: 因此,这些基本上是两个三重。一个是这个,莱比锡位于萨克森州,另一个是莱比锡位于德国。 00:59:41.090 EN: So, and if you remember the previous, so I hope, and some of you guys, Yaser et al, that are much more familiar than I am these days, please correct me if I'm stating something wrong here, but if you remember one of the previous slides, one of the first slides, had this information graphically represented. 中文初译: 所以,如果你记得前一张,我希望, 以及你们中有些人,Yaser等人, 比我现在熟悉得多, 如果我在这里说错了的话, 请纠正我, 但如果你还记得之前的幻灯片之一, 最初的幻灯片之一, 有这个信息用图形表示。 01:00:03.980 EN: Leipzig located in Saxony and then an arrow departing from Saxony located in Germany. And this is different from this statement. 中文初译: 莱比锡位于萨克森,然后箭从德国萨克森飞走。这与本声明不同。 01:00:15.900 EN: So this is Leipzig located in Saxony and Leipzig located in Germany. Versus the graphical representation was Leipzig located in Saxony, it's the same, and then Saxony located in Germany. 中文初译: 这就是莱比锡位于萨克森和列比锡位于德国的莱比锡。维苏斯的图形代表是莱比锡位于萨克森,它是一样的,然后萨克森位于德国。 01:00:30.620 EN: So the information is not exactly the same actually, but my point is that they are slightly different representations or serializations of this information. 中文初译: 所以信息实际上并不完全相同, 但我要说的是,它们只是 略有不同的信息表达方式或序列。 01:00:47.200 EN: And then this is a classic, we change here the property, Leipzig has mayor work at you. So these are yet another technique to shorten. 中文初译: 这是经典的,我们在这里改变财产, 莱比锡有市长在为您工作。所以这些是另一种缩短技术。 01:00:59.760 EN: Then about data types, these are already mentioned a little bit, so more as formal model to describe how different strings in the lexical domain are mapped through data type mapping to a range. 中文初译: 然后,关于数据类型, 这些数据已经略微被提及了一点, 所以更多的是正式模型 来描述如何通过数据类型绘图 绘制出不同字串的词组到一个范围。 01:01:20.460 EN: So as I mentioned, you can have literals in strings, 3.14, 3.14000, plus 03.14. They are all different strings, obviously, but by interpreting them as doubles, for instance, or decimals, you map these to the same range, 3.14. 中文初译: 因此,正如我所提到的,你可以有字串的字句,3.14,3.14000,加上03.14。它们都是不同的字符串,很明显,但是通过将它们解释成双倍,比如小数点或小数点,你就可以把它们映射到相同的范围,3.14。 01:01:46.200 EN: So this is the idea of data type mapping from a lexical domain to a range. And of course, depending on that data type mapping, if you use an xsd decimal data type for the mapping, then this mapping to the range is different, because if you type this as 3.14 string, then these three strings would in fact have a different range. 中文初译: 这就是从词汇域到范围的数据类型映射的概念。当然,根据数据类型映射,如果在映射中使用 xsd 小数点数据类型,那么这个映射到范围是不同的, 因为如果你输入3. 14 字符串, 那么这三个字符串实际上会有不同的范围。 01:02:14.620 EN: So they map to the same range only because you type these three strings using xsd decimal. And also, in this case, you have some changes. 中文初译: 所以它们映射到相同的区域, 仅仅因为您使用 xsd 小数点输入这三个字符串。 在这种情况下, 您也会有一些更改 。 01:02:27.800 EN: So 100.00 as a decimal would then map to a range. So it's sort of a canonicalization of these input strings to the canonical representation in the range domain, in the range space. 中文初译: 所以十进制为100.00,然后映射到一个范围。所以它是一种将这些输入字符串的卡通化, 在范围域、 范围空间的卡通表达面上。 01:02:47.280 EN: So 100.00 would then map as a decimal to 100. But as a string, it would still be mapped to 100.00. And that is, of course, implication against systems for sorting and querying also and so on. 中文初译: 也就是说,100.00将映射成小数到100,但作为字符串,它仍被映射为100.00。这当然意味着与分类和查询系统相悖。 01:03:13.800 EN: Good. Again, a little bit more about data types. So far, we had literals were untyped, treated as strings. So as I mentioned, sorting as strings are sorted, of course, differently than numbers. 中文初译: 。 再次,关于数据类型再多一点。 到目前为止, 我们的字典没有被打字, 被当作字符串处理。 因此正如我所提到的, 将字符串分类当然不同于数字。 01:03:31.780 EN: And typing is important because it allows sort of a semantic interpretation of values. And in RDF, data types are also URIs. 中文初译: 键入很重要,因为它允许对值进行某种语义解释。在“RDF”中,数据类型也是“URIs”。 01:03:45.340 EN: So you saw xsd string or xsd decimal in the previous line. There are actually also URIs. So xsd here is a prefix of a namespace in the schema data type. 中文初译: 所以,您在前一行中看到了 xsd 字符串或 xsd 小数点。 实际上还有“ URIs ” 。 因此, xsd 在 schema 数据类型中是命名空间的前缀 。 01:04:02.900 EN: So the semantic web technology stack and RDF, they're essentially built on xsd. And you probably remember the layered cake in the previous slide. 中文初译: 所以语义网络技术堆和“RDF” 基本上是建立在 xsd 之上的。你可能还记得上一张幻灯片中的分层蛋糕。 01:04:17.580 EN: You had, at the very bottom, you had xsd. So RDF builds on these technologies, URIs and xsd. But you can also introduce your own data types. 中文初译: 在底部,你曾经做过xsd。所以“RDF”建立在这些技术上,“URIs”和 xsd。但也可以介绍自己的数据类型。 01:04:32.520 EN: It's a flexible mechanism to introduce also data types for space, coordinates, or even units, and whatever you need, which are not in the xsd space, namespace. 中文初译: 这是一个灵活的机制,可以引入数据类型 用于空间,坐标,甚至单位, 以及任何你需要的, 这些不在 xsd 空间,命名空间。 01:04:47.560 EN: But you can introduce them and define them. And also RDF has two data types, XML literal and HTML, which are used to encode accordingly HTML and XML fragments. 中文初译: 但您可以介绍和定义它们。“RDF”还有两种数据类型,即“XML”字义和HTML,用来据此编码 HTML 和 “XML” 碎片。 01:05:09.980 EN: So an example. So we had this example, Leipzig as a subject has two properties, longitude and latitude, and two objects, in this case literals, they are not typed here in the graph. 中文初译: 举个例子。我们举了这个例子, Leipzig作为一个主体有两个属性,一个是经度和纬度,另一个是两个对象,在这个例子中,字面上没有在图表中输入它们。 01:05:24.900 EN: So you can translate this now in turtle. We had here prefixes that we introduced to have a compact representation. 中文初译: 所以现在你可以把这个翻译成乌龟。我们在这里有前缀, 我们介绍这些前缀是为了有一个缩略语。 01:05:36.640 EN: So DBR prefix with a namespace geo and another namespace and xsd. So we had this as well. And then mentioned xsd as a prefix for data types. 中文初译: DBR 前缀带有命名空间地理和另一个命名空间和 xsd 。 所以我们也有这个。 然后提到 xsd 作为数据类型的前缀 。 01:05:53.320 EN: And you can encode this information here with, yeah, we saw this already, laps the subject, geo lat and geo long as the properties, and then the literals. 中文初译: 你可以用这个信息编码, 是的,我们已经看到这个了, 绕过主题, 地理拉特和地理,只要这些属性, 然后是字面意思。 01:06:06.400 EN: And interestingly enough, it's not exactly the same, I would say, the graph, visual graph representation and the turtle serialization here, because here we actually explicitly type these literals as floats, and this is not done in the graph, in the visual representation of the graph RDF data. 中文初译: 有趣的是,它并不完全相同, 我想说,图表,视觉图示 和海龟的序列在这里, 因为这里我们实际上明确地将这些字面 输入为浮标, 而这一点并没有在图表中, 在“RDF”图形数据的直观表示中做到。 01:06:32.420 EN: Yes, I already mentioned language declarations, and this is a bit hard to see. So you can, here literals can also be tagged with a language tag. 中文初译: 是的,我已经提到过语言声明,这很难看。所以你可以在这里用语言标签标出文字。 01:06:48.760 EN: And these are the non-typed literals. So for instance here Leipzig, and if you add this at sign and then the language code, this is international language codes, so you can state that this is actually sort of a German translation of this literal. 中文初译: 这些是非类型文字。例如莱比锡语,如果你在签名时加上这个语言代码, 然后是语言代码, 这是国际语言代码, 所以您可以声明, 这实际上是一种德文翻译。 01:07:13.500 EN: And you can do the same in Russian, I trust this is correct. So Russian Leipzig would be the sort of Russian translation of this literal. 中文初译: 你可以用俄语做同样的话,我相信这是对的。所以俄罗斯莱比锡语就是俄文翻译的文字。 01:07:26.060 EN: And that's quite useful because you can, it's quite a nice feature because you can translate, have for instance labels of your resources in various languages, and then the client can simply query for the Russian translation of the label for this resource. 中文初译: 这是非常有用的,因为你可以, 这是一个相当不错的功能 因为你可以翻译, 比如用多种语言给资源贴上标签, 然后客户可以简单地查询这个资源的标签的俄文译文。 01:07:49.560 EN: And then you would just get the Russian translation, and you can display this in a Russian application for Russians. 中文初译: 然后你就会得到俄文翻译, 你可以在俄罗斯的俄国人申请中显示这个。 01:07:59.460 EN: And that's quite a nice feature of RDF, I think, to include a translation mechanism for labels in this manner. Yes, and then what this is also a note that in RDF11 things will be changed. 中文初译: 我认为,这是RDF的一个非常不错的特征, 用这种方式加入一个标签翻译机制。是的,然后这个也是一个注释, 在 RDF11 中的东西将会被改变。 01:08:22.300 EN: So in RDF10 Leipzig without data type and Leipzig with data type string, they were actually treated differently. 中文初译: 所以在RDF10 Leipzig没有数据类型,Leipzig有数据类型字符串,他们实际上得到了不同的对待。 01:08:36.980 EN: And this has changed in RDF11 where if someone states Leipzig without a data type, it's interpreted directly, automatically as a string. 中文初译: 在RDF11中,这已经改变了 如果有人说Leipzig没有数据类型, 它会直接被解读为字符串。 01:08:54.140 EN: Leipzig DE, so with a language tag, I think is still treated as a different literal. Good. Then a few words about NRE relations. 中文初译: Leipzig De, 所以用语言标签, 我认为仍然被看作不同的字面。 很好。 然后讲几句关于 NRE 关系。 01:09:10.750 EN: This is known from, I think, database systems. Sometimes you have more complex statements than simply, what was it, Albert Einstein has discovered relativity theory, where you have a sort of one subject, one predicate, and one object. 中文初译: 我想,这是从数据库系统中知道的。有时你的声明比简单的、什么更复杂, 阿尔伯特·爱因斯坦发现了相对论, 在那里你有一个主题,一个谓词和一个对象。 01:09:33.630 EN: These are simply encodable in RDF. Sometimes you have more complex, and mostly a few more complicated statements, like for instance here, cooking with RDF. 中文初译: 这些只是“RDF”中的编码而已。有时你还有更复杂的语句,大多是一些更复杂的语句,例如在这里用“RDF”做饭。 01:09:44.810 EN: So you have a preparation of a mango chutney, and to prepare mango chutney you have a number of ingredients. So you have here two ingredients, but potentially more. 中文初译: 所以,你准备了芒果甜甜圈 准备了芒果甜甜圈 准备了芒果甜甜圈 你有一些成分。所以这里有两种成分,但可能更多。 01:09:56.990 EN: So you need 450 grams of green mango, and also a teaspoon of cayenne pepper. And as you can see, if you start to think, OK, how am I encoding this into triples, it's not entirely obvious. 中文初译: 所以,你需要450克的绿色芒果, 和一支辣椒茶匙。正如你所看到的,如果你开始思考, 好吧,我如何将这个编码成三倍, 这并非完全显而易见。 01:10:14.690 EN: So the first attempt would be this one, for instance, OK, plain vanilla RDF on what we have so far introduced. You take a subject that will be clearly your chutney, which has ingredients. 中文初译: 所以第一个尝试就是这个,比如说,OK, 平凡的香草“RDF ” 我们到目前为止介绍的东西。你选一个主题, 这个主题将很明显是你的甜甜圈, 它含有成分。 01:10:32.350 EN: So you have a property ingredient, has ingredient, and you reuse this property for all the ingredients. So you have here one ingredient would be 400 grams green mango, that's the object, and you encode this as a literal, since it's not an identified resource, 400 grams of green mango, and also one tablespoon cayenne pepper, again, as a literal. 中文初译: 所以,你有一个属性成分,有成分, 你重新使用这个属性来配制所有元素。 所以你在这里,一个成分是400克绿色芒果, 这就是目标, 你把它编码成一个字, 因为它不是一个识别的资源, 400克绿色芒果, 还有一个茶匙辣椒,再次,作为一个字眼。 01:11:03.930 EN: There is, if you note, there is a slight error here, this should be a semicolon, because, oh no, sorry, yeah, no, it's right. 中文初译: 如果你注意到,这里有个小错误, 这个应该是分号, 因为,哦,对不起,是的,不,这是正确的。 01:11:13.310 EN: Actually, this uses the same property and subject, so this is correct, so it's a list of objects separated by comma. Sorry, by comma. 中文初译: 实际上,它使用相同的属性和主题, 所以这是对的, 所以它是用逗号分隔的对象列表。 抱歉, 逗号分隔 。 01:11:29.410 EN: But this is not so satisfying, how do we encode this information, right, because 400 grams of green mango and one tablespoon of cayenne pepper, yeah, there's more information than you would want to have in an atomic resource, right, or in an atomic literal. 中文初译: 但这不是那么满意, 我们如何编码这个信息, 因为400克的绿色芒果 和一桌辣椒, 是的,有比你想要的信息更多 在原子资源,正确的, 或者在原子字面上。 01:11:53.190 EN: So you would have to want to have 400 grams sort of separate from green mango, because it could be 400 grams of something else, so you could reuse 400 grams as a literal and linking it to something else. 中文初译: 所以,你必须想要有400克 与绿色芒果分离, 因为这可能是400克 其他的东西, 这样你可以重新使用 400克 作为字面上的, 并将它与其他的东西联系起来。 01:12:08.010 EN: So green mango may be better, this could be a resource, and you might want to use this in another ingredient with another amount. 中文初译: 所以绿色芒果可能更好, 这可能是资源, 你也许想把这个用在 另一种配料中,再加一个量。 01:12:18.250 EN: So this is not so elegant, also because for system it would be hard to, for instance, ask for things like, okay, give me all the recipes that have green mango, or give me all the recipes that have cayenne pepper, now independently on the quantities. 中文初译: 所以这不是那么优雅, 也因为对于系统来说, 很难要求 比如说,好吧,给我所有的配方 有绿色芒果, 或者给我所有配方 有辣椒的配方, 现在独立于数量。 01:12:38.990 EN: And this would be, yeah, harder to solve such an answers to query than it would be if you have modeled this a little bit better. 中文初译: 而这将会是,是的, 更难解决这样一个答案 查询比它会比较困难,如果你已经 模型这个稍微好一点。 01:12:52.970 EN: It's not impossible with, I guess, full text search, but it's not as elegant. So another variant, so the same sentence here, another attempt would be, okay, we separate, we separate 450 grams from the green mango and one tablespoon from the cayenne pepper, and you say, okay, we have the chutney, we have an ingredient, that's the green mango, and then we have an amount, 450 grams. 中文初译: 全文搜索并不是不可能的,我想, 但是它不是那么优雅。所以另一个变式, 所以这里的句子一样, 另一种尝试是,好吧,我们分开, 我们从绿色芒果中分离450克, 从辣椒中分离出一张桌子, 然后你说,好吧,我们有巧克力,我们有一个成分, 这就是绿色芒果, 然后我们有一个数量,450克。 01:13:20.370 EN: Then we have, however, another ingredient, cayenne pepper, and then we have an amount, one tablespoon. That's even perhaps worse, because now you don't have an obvious unambiguous assignment anymore between the ingredients and the amounts. 中文初译: 然而,我们有另一种成分,辣椒, 然后我们有一个数量,一个面板。这也许更糟糕了,因为现在你不再对成分和数量有明显的明确区分。 01:13:36.610 EN: So remember that contrary to XML, where you have an order, here you don't really have an order. So RDF data are triples, and this is a set, so there is no ordering. 中文初译: 所以请记住,与有命令的“XML”相反, 这里你并没有真正的命令。 所以“RDF”数据是三倍, 这是一套, 所以没有命令 。 01:13:56.010 EN: And therefore, it would be completely equivalent if you would say chutney, ingredient mango, ingredient cayenne pepper, amount one tablespoon, amount 450 grams. 中文初译: 因此,如果你说巧克力, 配料芒果, 辣椒成分, 一个面粉,450克。 01:14:08.510 EN: So these two RDF sets of RDF statements, triples would be completely equivalent. And so you don't have an assignment, so which one is now 450 grams? 中文初译: 所以这两组“RDF”的“RDF”声明, 3倍是完全等同的。所以你没有任务, 那么现在哪个是450克? 01:14:23.410 EN: Is it green mango, or is it cayenne pepper? And of course, it makes a huge difference. If you use 400 grams of cayenne pepper and only one tablespoon of green mango, you're not going to get the chutney. So this is bad here. 中文初译: 这是绿色芒果,还是辣椒?当然,它有巨大的不同。如果你使用400克辣椒和只用一块绿芒果桌,你就不会得到甜甜圈了。所以这里情况很糟。 01:14:38.950 EN: So that's why we introduced this NRE relation. So in a way, what you have here with the mango chutney and the green mango and the amount, these are in themselves ternary relations. 中文初译: 所以这就是为什么我们引入了NRE关系的原因。从某种意义上说,你这里的芒果巧克力和绿色芒果 以及数量, 这些是他们自己与世世代代的关系。 01:14:55.830 EN: And in databases, you have this. You just extend the tables with additional columns. But in RDF, you can't do this as easily because the basic unit, they are triples. 中文初译: 数据库里有这个。你只要用额外的列来扩展表格。但是在“RDF”中,你不能这么容易做到这一点,因为基本单位是三倍。 01:15:09.790 EN: So it's a structure of three things, subject, predicate, and object. So we need to have another mechanism to encode these more complex relations. 中文初译: 所以这是由三件事组成的结构, 主题、谓词和对象。 所以我们需要另外一种机制来编码这些更复杂的关系。 01:15:18.990 EN: And the solution is to introduce these helper nodes, which are essentially these blank nodes that I mentioned before. 中文初译: 解决办法是引入这些助手节点, 这些节点基本上是我之前提到的空白节点。 01:15:25.830 EN: Well, actually, blank nodes are one possibility, and I'm coming to this. You can also identify these nodes. These are helper nodes more generally. 中文初译: 事实上,空节点是一种可能性。我正要说到这一点。你也可以识别这些节点。这些节点一般都是辅助节点。 01:15:34.130 EN: So what you do here is you have the chutney as a subject, the chutney as an ingredient, and then you introduce a helper node here. 中文初译: 所以,你在这里做的是 你有一个切特尼作为主体, 切特尼作为一个成分, 然后你在这里介绍一个助手节点。 01:15:44.630 EN: So in this case, it's actually identified. So it has a URI, chutney ingredient one. And this chutney ingredient one has now two properties. 中文初译: 因此,在这个案例中,它实际上是被识别的。 所以它有一个“URI”, 酸奶成分之一。 而这个巧克力成分之一现在有两个属性。 01:15:54.990 EN: One is the ingredient and one is the amount. This is the same as we had before. So we had also split, remember, the ingredients and amounts. 中文初译: 一个是成分,一个是数量。这是和我们以前一样的。 所以我们也分开了,记住, 成分和数量。 01:16:04.810 EN: But the difference is that now they are not related to the chutney, but they are related to this ingredient amount. 中文初译: 但两者的区别在于,现在它们与巧克力无关, 但它们与这个成分含量有关。 01:16:10.770 EN: They are related to this helper node, chutney ingredient one, which therefore allows for a clear distinction between the ingredients. 中文初译: 它们与这个辅助节点有关,即巧克力成分1,因此可以明确区分这些成分。 01:16:21.230 EN: So the clear assignment of amounts with ingredient in the recipe for chutney. And this has a different encoding, of course, here in interval. 中文初译: 所以在切特尼食谱中清晰地分配了含有成分的量。 当然,这有一个不同的编码, 间隔在这里。 01:16:33.210 EN: So you have the chutney. It has an ingredient. And then you introduce here an identified resource, which is this helper node, chutney ingredient one. 中文初译: 所以,你拥有了切特尼。它有一个成分。然后在这里介绍一个识别的资源, 也就是这个助手节点, 切特尼成分之一。 01:16:44.150 EN: And then you describe the chutney ingredient one as having an ingredient. That will be green mango. And this is an actual resource here, identified. 中文初译: 然后,你将一个切特尼成分描述为有一个成分。那就是绿色芒果。这是这里的一个实际资源,被识别出来。 01:16:54.050 EN: And the amount is literal, 450 grams. Now you could say the same thing about this 450 grams and say, oh, OK, but this is a more complex literal. 中文初译: 数量是字面的,450克。现在你可以对这450克说同样的话,然后说,哦,好吧,但是这是一个更复杂的字面。 01:17:05.190 EN: I would like to have this as a value, 450 and the grams as a unit. So you could also introduce here, again, a helper node instead of this literal and separate and say, OK, here you have a chutney ingredient one amount as a helper node. 中文初译: 我想用这个值,450克作为单位。所以你也可以在这里再次介绍一个助手节点, 而不是这个字面上的和分开的节点, 然后说,好吧,这里你有一个切特尼成分 一个数量作为一个助手节点。 01:17:23.150 EN: And then it has a value 450. And then a unit has unit to a resource for the unit grams, which would be even more elegant so that you can parse this or query this more elegantly. 中文初译: 然后它就有一个450的值。然后一个单位就有一个单位 来作为单位克的资源, 这个单位会更优雅一些, 这样你就可以分析这个或者更优雅地查询这个。 01:17:42.000 EN: And here is where the blank nodes come into play. So we had here before in the previous slide a named resource for this helper node. 中文初译: 这里就是空白节点起作用的地方。 所以我们在前一张幻灯片中, 曾经有一个名为这个助手节点的资源 。 01:17:54.640 EN: In this case, that's not really necessary to explicitly identify these helper nodes. It's just you're not going to address this independently from the chutney recipe. 中文初译: 在此情况下, 并不真的有必要明确识别这些帮手节点。 只是你无法独立于切特尼配方来解决这个问题。 01:18:09.620 EN: And so it's always a node linked to this chutney recipe as an ingredient. So that's why you could introduce, you could use blank nodes, which are nodes, regular nodes, but they don't have a name. 中文初译: 所以它总是与这个古特尼配方的节点相连。所以这就是为什么你可以介绍, 你可以使用空白节点, 它们是节点,常规节点, 但是它们没有名字。 01:18:25.460 EN: And this has a slightly different encoding. So again, there is no name here, no URI. Otherwise, the graph is the same. 中文初译: 这个编码稍有不同。所以这里没有名字,也没有“URI ” 。 否则,图表是相同的。 01:18:34.700 EN: And this is slightly encoded differently, you see here in TURTL. Using, so this is TURTL grammar here, syntax, it uses this underscore colon ed1. 中文初译: 这是略微不同的编码, 您可以在 TURTL 中看到。 使用, 所以这里是 TURTL 语法, 语法, 它使用这个强调冒号的编辑 1 。 01:18:49.360 EN: This is introduced automatically by the parsers, by the systems. You don't have to encode this yourself. So chutney has ingredient, but otherwise it's the same thing. 中文初译: 这是由解剖师和系统自动引入的。 你不必自己编码它。 所以Chutney有成分, 否则就是一样的了。 01:19:02.000 EN: So you link this object to this subject here, and the ingredient is just not named now. But then the ingredient has these two properties, ingredient green mango and amount, again, the 450 grams. 中文初译: 所以您将这个对象与此主题链接, 成分现在还没有命名。 但是元素有两种特性, 成分绿色芒果和数量, 也就是450克。 01:19:17.720 EN: And this can be shortened elegantly, I think, with these squared brackets. And you can see here again, very human readable, human friendly syntax. 中文初译: 我认为,这可以用这些方括号来简化。你可以在这里再次看到,非常人性易读, 人类友好的语法。 01:19:30.860 EN: You can simply, so there is this curly square brackets, they introduce a blank node as a helper node here to state that chutney has an ingredient. 中文初译: 你可以简单地说,所以有这个卷轴方括号, 他们在这里输入一个空白节点作为辅助节点 来说明Chutney有一个成分。 01:19:43.600 EN: And this is this complex thing that is not named that has an ingredient green mango and amount 450 grams. So they are equivalent, but this is shortened, abbreviated, and a little bit more elegant to read, also because there's no information carried by this node identifier here. Well, it's not a URI, remember. 中文初译: 这是这个没有命名的复杂事物, 配有绿色芒果和450克的成分。 所以它们相等, 但是它被缩短了,缩略了, 读起来更优雅一些, 也因为这里没有通过节点识别符传递信息。 嗯, 它不是“ URI ” , 请记住 。 01:20:12.460 EN: So lists, how much do you have on time? Okay, still a few minutes. So let me just briefly also introduce lists here. 中文初译: 单子上,你有多少时间?好吧,还有几分钟。请允许我在此简短地介绍一下清单。 01:20:19.700 EN: So that's another concept of RDF, because as I said before, RDF doesn't have a stable sorting, it's a set. So sometimes you actually want to sort stably something. 中文初译: 这是另一个“RDF”的概念,因为正如我之前所说, “RDF”没有稳定的分类, 它是一个组合。所以有时你真的想对一些东西进行细化。 01:20:35.100 EN: I don't know, numbers, or you know, you have perhaps people in a group, you want to in a research group, you want to state that these are sort of a set, a collection of, or a container here of people that are members of some group. 中文初译: 我不知道,数字,或者你知道, 你也许有一群人, 你想加入一个研究组,你想声明 这些是一组人, 一个集合的, 或一个集装箱 这里的人是某个群体的成员。 01:20:57.380 EN: And you want to state this explicitly in RDF. So these are what these containers and collections are for. And there is this distinction between container which allow to add new elements. 中文初译: 您想要在“ RDF ” 中明确说明这一点。 所以这些容器和收藏是用来做这些的。 在容器之间有这种区别,可以添加新的元素。 01:21:10.260 EN: So if there is a container defined in some RDF data, you are allowed to add new elements to these containers in your own RDF data. 中文初译: 因此,如果有些“RDF”数据中定义了一个容器,那么允许您在自己的“RDF”数据中为这些容器添加新的元素。 01:21:21.340 EN: And they are, they can be ordered and unordered container types. And we will talk, I will say a little bit more about this in the next slides. 中文初译: 它们是,它们可以被订购和无订单的集装箱类型。我们将在下一张幻灯片中谈论,我将多说一点。 01:21:32.500 EN: And then you have collections, these are ordered lists. And two ordered lists, you are not allowed to add additional elements. 中文初译: 然后,你有收藏品,这些是定购单。两个定购单,不允许添加额外的元素。 01:21:41.860 EN: And there is a mechanism that will briefly describe how this is done. So the containers, there are three types of containers. 中文初译: 有一种机制将简要描述如何做到这一点。因此,集装箱有三种类型的集装箱。 01:21:54.240 EN: The RDF sequence here, that's an ordered list as a sequence. The interpretation is an ordered list as a sequence. 中文初译: RDF 序列, 这是按顺序排列的列表。 解释是按顺序排列的列表 。 01:22:05.280 EN: The bag is an unordered set. And alt, which is not so common in my opinion, I haven't seen it much, is a set of alternatives. 中文初译: 包是一套没有顺序的套装。而替代物, 在我看来并不常见,我还没有看到太多, 是一套替代物。 01:22:18.180 EN: So usually only one of the list elements is, or container elements is, is relevant. So these are the containers, three types of containers as the sequence, the bag, and alternatives. 中文初译: 因此,清单中通常只有一个要素是相关的,或集装箱要素是相关的。因此,这些要素是集装箱、三种集装箱序列、袋和替代物。 01:22:37.830 EN: So this is an example for how to, you can encode in graphically visually represented here a sequence. So you have here AKSW, which is a research group in Leipzig, has a number of members. And we introduce here a helper node, a blank node, to encode these relations to the members. 中文初译: 因此,这是一个如何用图形化的视觉来编码一个序列的示例。 所以在这里您可以使用AKSW(AKSW)这个在莱比锡的一个研究小组拥有若干成员。 我们在此介绍一个助手节点, 一个空白节点, 来将这种关系编码给各成员。 01:23:02.370 EN: So this is a sequence and it's ordered. So we are using here a dedicated property here in RDF, in the RDF namespace. 中文初译: 这是一个序列,它已经订购了。所以我们在这里使用一个专用财产 在“RDF”,在“RDF”的命名空间。 01:23:12.570 EN: Underscore one, two, three, four, that's sort of the stable sorting, remember sequences are sorted. And then you have these are the properties to the objects, which are resources here. And these are identified people. 中文初译: 一、二、三、四,这是一种稳定的排序, 记住序列是分解的。 然后这些是物体的属性, 这些是这里的资源。 这些是被识别的人。 01:23:27.370 EN: So Søren, Jens, Axel and Michael would be members of the AKSV group. And this is, this node here is actually, or this resource here is a sequence of people, members of this group. 中文初译: Soren, Jens, Axel和Michael 将成为AKSV组的成员。这就是,这里的节点是, 或者这个资源是一组人, 这个组的成员。 01:23:47.430 EN: Then we have also collections. Remember we have collections and containers, and the containers have these three different types, sequence, alternatives and bags. 中文初译: 然后我们还有收藏品。记住我们有收藏品和容器, 集装箱有三种不同的类型、序列、替代物和袋子。 01:24:02.450 EN: And collections, and the difference to collections is that sequence, containers could be extended with additional elements. 中文初译: 收集,与收集的不同之处在于 序列,容器可以扩大 附加元素。 01:24:11.590 EN: So someone else could say, okay, I have an additional member of the AKSV group and I'm adding this to this collection. 中文初译: 所以别人可以说,好吧,我还有一个AKSV组的成员 我要把这个添加到这个收藏中。 01:24:21.370 EN: And now lists are closed. So someone else cannot add additional members to the list. And this is done like this, that you have this mechanism of, the idea here is recursive partition of lists into head elements and possibly empty rest lists. So what does this mean? 中文初译: 现在列表已被关闭 。 所以其他人不能在列表中添加额外的成员 。 这样就完成了, 您有这样的机制, 这里的想法是将列表的递归分割成头部元素, 并可能空白的休息列表 。 这意味着什么 ? 01:24:45.990 EN: So you have the group AKSW has a number of group leaders, and that's a stable set. That's a finite set. So that's a well-defined set. 中文初译: 所以AKSW组有多个团体领袖, 这是一套稳定的组合。这是一套有限的组合。所以这是一套定义明确的组合。 01:24:56.050 EN: Nobody else can add additional leaders. Søren, Jens and Axel. And how this is done, the list is that you have a first element, that's Søren, and then the rest of the list. 中文初译: 没有人可以增加更多的领导人。瑟伦,延斯和阿克塞尔。如何做到这一点,名单是:你有一个第一个要素,那就是索伦,然后是名单的其余部分。 01:25:08.190 EN: So this helper node here links then on one hand to the first element, that's Søren, and to the rest of the list with this RDF rest property, which really is all the rest. 中文初译: 所以这个助手节点 将第一个元素,即瑟伦, 和其余的列表链接到这里, 加上“RDF”的休息财产, 这些都是其他所有的东西。 01:25:21.270 EN: And then of course in the rest, you have again a helper node, which has a relation to the first and the rest. So the first is always then in this case a person, and the rest is the remaining of the list. 中文初译: 当然,在剩下的部分,你还有一个助手节点, 它与第一个和其余的有关系。所以第一个节点总是一个人, 其余的则是名单的剩余部分。 01:25:36.010 EN: And then you have here third person, again linked to a helper node. You have here property first, which links to another member of the group leaders at the AKSW group, and then the rest of the list. 中文初译: 然后这里有第三个人,又与一个助手节点相连。你在这里先拥有财产,它与AKSW集团的另一个团体领导人有关,然后是名单的其余部分。 01:25:51.790 EN: And as I said before here, the rest of the list could be possibly empty. So when the list is terminated, you have the rest points to a special resource in RDF, called RDF nil, and that's a null node, a nil node, and this is an indication that this list is closed. So there is no rest, essentially. 中文初译: 正如我先前在这里说过的那样, 列表的其余部分可能是空的。 所以当列表终止时, 您将其余部分指向一个名为“ RDF”的“ RDF” 无的特殊资源, 这是个空节点, 一个无节点, 这是显示这个列表已经关闭。 因此基本上没有休息了 。 01:26:20.400 EN: Maybe take a look at this again at home, because it's a little bit complicated maybe at first to figure this one out. 中文初译: 也许在家里再看看这个, 因为它有点复杂,也许一开始就想出了这个。 01:26:27.880 EN: But if you take a second look, you can see the difference between sequence and lists here, and how one is open and the other one is explicitly closed. 中文初译: 但如果你再看一眼, 你可以看到顺序和列表之间的差别, 以及一个是如何打开的,另一个是明确关闭的。 01:26:41.260 EN: So this can be encoded very nicely also in turtle, so very shorthand way to represent lists with this parentheses here. So you just list the members here of the AKS, group leaders of the AKSW group as resources, but this is actually interpreted as a list with all this RDF first rest and RDF nil. 中文初译: 所以这也可以在海龟中进行非常精美的编码, 因此非常简略地代表列表, 并在此括号内 。 所以您只需列出 AKS 的成员, AKS SW 组组长, 作为资源, 但实际上人们会将它解释为所有“ RDF” 第一休息和“ RDF”零的列表 。 01:27:20.760 EN: This compact representation of this list is actually represented internally as a list. Good. Summary. Let me see. Okay, we're almost done. 中文初译: 这份清单的缩略语实际上以内部列表形式表示。好,摘要,让我看看。好的,我们快完成了。 01:27:38.200 EN: So we have here with RDF a standard, which can be used to store, but in particular exchange data, represent data. 中文初译: 因此,我们这里有“RDF”标准, 可用于存储, 但特别是交换数据, 代表数据。 01:27:49.040 EN: It's a data model and it's textual, so you can easily exchange this data also across systems. It's not binary, so there are lots of parsers that can easily parse. 中文初译: 这是一个数据模型,它是文字的, 所以你可以很容易地在系统之间交换这些数据。它不是二进制的, 所以有很多解剖师可以容易地分析。 01:28:02.920 EN: You can also take simply a look at it in Notepad or whatever text editor. So that's quite handy, and you can exchange, of course, this data very easily over HTTP protocols. 中文初译: 您也可以在 Notepad 或任何文本编辑器中简单查看它。 这样非常方便, 当然, 您可以通过 HTTP 协议很容易地交换这些数据 。 01:28:17.780 EN: Yeah, so I'm not sure what else there is to say about this. So enables syntax independent representation of distributed information in graph-based data model. 中文初译: 是的,所以我不确定还有什么要说的。因此可以让语法独立代表 以图表为基础的数据模型中传播的信息。 01:28:30.620 EN: Yeah, that's a complicated sentence to simply say that the RDF data model is a data model, and you have a number of syntaxes for serializing this data model into strings. 中文初译: 是啊,这是一个复杂的句子 简单地说“RDF”数据模型是一个数据模型, 你有一些语法把这个数据模型序列成字符串。 01:28:47.360 EN: As we saw, turtle is one, but there are a lot more, a few more, which you will see in the next lecture. So this data model is indeed syntax independent from a particular serialization. 中文初译: 正如我们所看到的,海龟是一只,但还有更多,更多的,你会在下一次演讲中看到。因此这个数据模型的语法与特定的序列化无关。 01:29:07.180 EN: And distributed information, yeah, so you can have data in graph-based data model. RDF is a graph-based data model, but this data can also be distributed in systems. 中文初译: 以及分布式信息,是的,所以你可以在图表数据模型中拥有数据。“RDF”是一个基于图表的数据模型,但这些数据也可以在系统中传播。 01:29:24.600 EN: You can, since we use globally unambiguous identifiers, URIs, you can say, so one system can state something about Joe Biden, another system can state something else about Joe Biden. 中文初译: 你可以,因为我们使用全球范围 毫不含糊的识别资料, “URIs”,你可以说, 所以一个系统可以说明乔・拜登的一些情况, 另一个系统可以说明乔・拜登的其他情况。 01:29:39.700 EN: And so you can see that the information about Joe Biden can be distributed, and thanks to using the same identifiers, you can easily integrate this information from a distributed system. 中文初译: 这样你就能看到关于Joe Biden的信息可以被传播, 并且由于使用同样的识别资料, 你可以很容易地从分布式系统中整合这些信息。 01:29:54.820 EN: And that's quite interesting feature of RDF in contrast to, for instance, relational databases, which in general don't have these global identifiers, at least not within the database as a feature. 中文初译: 这是“RDF”相当有趣的特征, 与(比如)关系数据库不同。 一般来说,它没有这些全球识别资料, 至少不在数据库里作为特性。 01:30:06.720 EN: Of course, you can encode URIs, also include them in tables, but the identifiers, foreign keys and so on, in relational database management systems, they are local identifiers to the particular database. 中文初译: 当然,您也可以将“URIs”编码为“URIs”,同时也将其包括在表格中,但识别符号、外国密钥等等,在相关数据库管理系统中,它们是特定数据库的本地识别符号。 01:30:24.860 EN: And if you want to share information, it's not, with other systems, in a distributed system, it's not entirely obvious sometimes to identify the same as being the same. 中文初译: 如果你想共享信息, 它不是与其他系统一样, 在分布式系统中, 有时并不完全明显地识别相同。 01:30:40.770 EN: Yeah, so the last point here is that, so RDF is very individual oriented, resource-centric, so you can state a lot of things about Chutney and its ingredients, but you cannot organize this data in a schema. 中文初译: 所以这里的最后一点是, 所以“RDF” 是一个非常个人化的, 以资源为中心的, 所以你可以说很多关于Chutney及其成分的事情, 但你不能用一个计划来组织这些数据。 01:31:02.500 EN: You cannot say that, for instance, XYZ are people, PQR are books. And for this, there is an additional technology called RDF schema, and also on top of that, things like the web ontology language. 中文初译: 你不能说,例如,XYZ是人,PQR是书。为此,还有一种叫“RDF”的图案技术, 以及网络本体语言等。 01:31:23.430 EN: And these languages, they allow you to organize a little bit of RDF data into classes and property types, and you will see more about this in a future lecture. 中文初译: 这些语言,它们允许你 将一些“RDF”数据 整理成分类和属性类型, 你会在未来的演讲中看到更多关于这一点的信息。 01:31:36.610 EN: Yeah, a few news about RDF 1.1, what has changed. You can take a look at this. It has been a recommendation since February 2014, and there is a complex spec you can read online, and a few things have, of course, changed. 中文初译: 有关RDF 1. 1. 1 的几则消息, 改变了什么。 您可以看看这个。 从2014年2月起, 这是一条建议, 而且有一个复杂的特征, 你可以在网上阅读, 当然有些东西已经改变。 01:31:52.850 EN: And then in the end, there are some tasks and mini projects. I think I leave this now as the homework, and I'm not entirely sure, maybe you can chime in. 中文初译: 最后,还有一些任务和小型项目。我想现在我把这个留给做作业, 我不完全确定,也许你可以进来。 01:32:04.570 EN: Who is leading the exercises here this semester? That should be me, Marcus. That should be you, okay. So how are you organizing this? 中文初译: 这学期谁来领导训练? 01:32:15.470 EN: Are you starting a little bit later or? Basically, it is next week. Yeah, the sheet is already also uploaded to Stude IP, so students should already have access to it. 中文初译: 您是在晚一点开始还是? 基本上, 这是下周。 是的, 床单已经被上传到 Stude IP 上, 所以学生应该已经可以访问它了 。 01:32:25.810 EN: And I also pasted the link for the WebEx meeting as well in the chat. Great. And then you're going through these exercises here or? 中文初译: 我还把WebEx会议的链接贴在聊天上,很好。然后你在这里还是在这里做这些练习? 01:32:35.750 EN: We can take a look. So there is a bit of a kind of review phase. So we start with a bit with the concepts that also were mentioned there, and a bit of a toy example for people to get their hands dirty a bit with RDF. 中文初译: 我们可以看一看。所以有一个有点像审查的阶段。 所以我们从一些概念开始, 也提到了这些概念, 和一个玩具的例子, 人们用“RDF”把自己的手弄脏。 01:32:50.570 EN: Great. So yeah, take a look at these examples or exercises here. There are even solutions, so it's very easy, I think, to follow this. 中文初译: 很好,所以请看看这些例子或练习。这里甚至有解决办法, 所以,我想很容易遵循这一点。 01:33:01.630 EN: And that was it for today. So maybe you have any questions. I know that we are slightly over time, but I can sit here a little bit longer. 中文初译: 今天就到此为止了。所以也许你有任何疑问。我知道我们时间有点长, 但我可以坐在这里多坐一会儿。 01:33:12.030 EN: So if you have questions, please stay with me. You can also ask in the chat or just by raising hands. Otherwise, thank you very much for joining today. 中文初译: 所以,如果您有问题,请留在我身边。您也可以在聊天或举手询问。否则,非常感谢您今天的加入。 01:33:24.030 EN: I hope it was a clear introduction to RDF, very important piece of technology. And take a look at it still at home before next week because the basics of RDF are important to keep in mind for the remaining lecture material. 中文初译: 我希望这是“RDF”的一个清晰的介绍,它是一个非常重要的技术。看看下周之前仍然留在家里,因为“RDF”的基本内容对于其余的教学材料很重要。 01:33:40.010 EN: So if you are a simulated weldist material, then you have an easier time also in subsequent lectures to follow what Søren or some colleagues will tell you. 中文初译: 所以,如果你是模拟的焊工材料, 那么在以后的讲座上,你也有比较容易的时间 来听索伦或一些同事会告诉你的话。 01:33:53.330 EN: So thank you very much for joining. Hope you have a good day, remaining week. And yeah, I think next week we'll be back soon. So bye for now. 中文初译: 非常感谢你们参加。希望你们今天过得愉快,剩下的一周。是的,我想下周我们会很快回来。现在再见。 01:34:03.590 EN: And as I said, I will stay a little bit longer if you have any questions. Thank you. 中文初译: 正如我所说,如果你有任何问题,我会多待一会儿。谢谢。