A Programmable Web/Chapter 1 > 공지사항

• 목록
• 답변
• 글쓰기
• 게시판 리스트 옵션
  • 수정
  • 삭제

In case you are like most individuals I know (and, since you’re reading this e-book, you probably are-at least in this respect), you utilize the online. So much. In actual fact, in my very own private case, the vast majority of my days are spent studying or scanning web pages-a scroll by my webmail shopper to talk with associates and colleagues, a weblog or two to catch up on the information of the day, a dozen short articles, a flotilla of Google queries, and the constant flip to Wikipedia for a stray fact to answer a nagging query. All fine and good, in fact; certainly, nigh indispensable. And but, it is sobering to think that little over a decade ago none of this existed. Email had its personal specialized functions, weblogs had but to be invented, articles had been found on paper, Google was yet unborn, agreement signature and Wikipedia not even a distant twinkle in Larry Sanger’s eye.

And so, it is striking to consider-virtually shocking, the truth is-what the world may be like when our software turns to the web simply as continuously and casually as we do. Today, in fact, we will see the faint, future glimmers of such a world. There's software program that phones residence to find out if there’s an replace.There may be software where a part of its content material-the help pages, perhaps, or some type of catalog-is streamed over the net. There is software program that sends a copy of all your work to be stored on the internet. There is software specifically designed to help you navigate a sure sort of net web page. There is software program that consists of nothing but a certain type of net web page. There may be software program-the so-called "mashups"-that consists of a web web page combining info from two different web pages. And there is software that, utilizing "APIs," treats different internet sites as simply one other a part of the software infrastructure, another function it might call to get issues performed.

Our computers are so small and the online so great and vast that this final scenario looks as if part of an inescapable development. Why wouldn’t you depend upon different web sites each time you might, making their countless data and bountiful skills a seamless part of yours? And so, I believe, such uses will develop into more and more frequent till, sooner or later, your computer is as tethered to the online as you your self are actually. It is typically recommended that such a future is not possible, that making an online that other computers might use is the fantasy of some (reasonably unimaginative, I'd think) sci-fi novelist. That it might solely happen in a world of lumbering robots and artificial intelligence and machines that observe you round, barking orders whereas intermittently unsuccessfully making an attempt to steer you to buy a new pair of shoes. So it is maybe unsurprising that one of many critics who has expressed one thing like this view, Cory Doctorow, is in reality a quite imaginative sci-fi novelist (amongst a lot else).

Doctorow argues that any system acquire correct "metadata"-the type of machine-processable information that will be wanted to make this dream of computer systems-utilizing-the-Web come true-will run into seven inescapable problems: people lie, people are lazy, persons are stupid, individuals don’t know themselves, schemas aren’t neutral, metrics influence outcomes, and there’s greater than one way to describe one thing. Instead, Doctorow proposes that as a substitute of trying to get people to supply data, we should always as a substitute have a look at the information they produce by the way whereas doing other things (like how Google looks at the links people make when they write internet pages) and use that as an alternative. Doctorow is, after all, attacking a strawman. Utopian fantasies of sincere, complete, unbiased information about the whole lot are clearly unimaginable. But who was making an attempt for that anyway? The web is rarely completely trustworthy, complete, and unbiased-however it’s still fairly rattling useful. There’s no cause making an online for computer systems to make use of can’t be the identical way.

I need to say, nevertheless, the idea’s proponents do not escape culpability for these utopian perceptions. Lots of them have gone around talking concerning the "Semantic Web" in which our computer systems would finally be able to "machine understanding." Such a framing (among other factors) has attracted refugees from the struggling world of artificial intelligence, who have taken it as one other alternative to promote their life’s work. Instead of the "let’s just construct one thing that works" perspective that made the web (and the Internet) such a roaring success, they brought the formalizing mindset of mathematicians and the institutional buildings of teachers and protection contractors. They formed committees to kind working groups to put in writing drafts of ontologies that rigorously listed (in 100-web page Word paperwork) all possible things within the universe and the varied properties they could have, and so they spent hours in Talmudic debates over whether a washing machine was a kitchen appliance or a household cleansing device.