TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

The Future of Content Management

19 点作者 techdog超过 15 年前

3 条评论

9oliYQjP超过 15 年前
Actually, I beg to differ with the author. The problem I've come across with a lot of CMSes is that they have tried to make content so abstract as to make it cumbersome to do anything with it. The only thing that really exists with content is its concrete representation at any given point in time, whether it's HTML, a database table row, or an audio recording. You can't store abstract ideas, but that's precisely what CMSes are trying to do. All the content is broken down into primitive types. Structure and sequence are removed as much as possible. Then more complex content is assembled from these smaller pieces. The problem is, this seems like the right approach but it quickly breaks down.<p>For instance, I've dealt with CMSes that have tried to remove HTML from their core content only to be forced to store HTML strings in their data stores and use an editor like FCKEdit or TinyMCE in later versions of the product. The problem has never been the extra structure and sequence that HTML lends to the content. The problem has always been that it has been cumbersome to transform this structure and sequence. When someone wants to change their website so an article written on one HTML page is now spread across 5 pages, it shouldn't require them to have to jump through hoops to accomplish this. Or if somebody wants to convert a written article into a spoken recording, a CMS should be able to handle this.<p>It's alright to store content in its final publishable state of HTML, so long as you make it easy to transform that content into different representations. How you store the intermediary data does not matter. It's the transformations between final content states that are the key. But today's CMSes make it so difficult to transform content. Think about it. How straightforward is it to move a website from one CMS to another let alone from say, a website to a printed book? I honestly think the future of CMSes lies in something that looks a lot like Yahoo Pipes rather than some centralized repository enterprise battleship clunker (can you tell I've had to deal with some crappy CMSes)?
itgoon超过 15 年前
How long until someone posts that the "future of CMS" is actually a centralized index, with the data existing wherever it makes sense? Kind of a personal card catalog.<p>FWIW, that's exactly what I'm working on.
评论 #796396 未加载
Rickasaurus超过 15 年前
I've tried to solve this on my own and, while it seems simple, it's hugely complex. What I want is just a black box that I can stuff data into and have it be semantically tagged.