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。

Please admit defeat (HTTP WG)

41 点作者 Supermighty超过 10 年前

6 条评论

mike_hearn超过 10 年前
As far as I can tell everyone thinks HTTP&#x2F;2 is a good idea except one guy, whose HTTP cache software doesn&#x27;t even implement HTTP&#x2F;1.1 fully.<p>His reason for why HTTP&#x2F;2 sucks is that it&#x27;s not ambitious enough, yet when the WG chair says &quot;v2 isn&#x27;t perfect but we can always do a v3&quot; he also gets upset.<p>Apparently this guy&#x27;s hobby is boiling oceans. I think we can give it a rest. There&#x27;s no real controversy or drama here. There&#x27;s just one guy who is very vocal.
评论 #8882762 未加载
评论 #8882746 未加载
评论 #8882904 未加载
评论 #8882848 未加载
cnst超过 10 年前
I think HTTP&#x2F;2 should have opportunistic encryption support, as per <a href="https://lists.w3.org/Archives/Public/ietf-http-wg/2015JanMar/0106.html" rel="nofollow">https:&#x2F;&#x2F;lists.w3.org&#x2F;Archives&#x2F;Public&#x2F;ietf-http-wg&#x2F;2015JanMar...</a>.<p>As things stand, HTTP&#x2F;2 is not done in compliance with RFC 7258 &#x2F; BCP 188, because it doesn&#x27;t do anything for the independent sites that cannot deploy the `<a href="https://`" rel="nofollow">https:&#x2F;&#x2F;`</a> address scheme for compatibility reasons.<p>You basically either have to support the whole pre-1.2 TLS and forget about <a href="http://" rel="nofollow">http:&#x2F;&#x2F;</a>, or you cannot have any TLS at all.
评论 #8884772 未加载
stormbrew超过 10 年前
The worst thing about all of this is that there has never been a good reason to push through anything as HTTP&#x2F;2 at this time. The key innovation of spdy wasn&#x27;t any of the technical aspects of the protocol itself, but the fact that it was making good on the idea that protocol upgrade is a feasible prospect.<p>Formalizing and standardizing the means by which we can upgrade to spdy now, and later to http&#x2F;1.2 or &#x2F;2 (beyond the lipservice paid to the idea in the HTTP&#x2F;1.1 RFCs), would have done a great service to the evolution of the web. It also would have allowed a reasonable time for other alternatives to spdy to show up, rather than a standardization timeline that was so ridiculously narrow as to allow only one possible alternative to have enough experience in the wild to succeed.<p>To sum up: spdy should have been standardized as spdy, not http&#x2F;2, and the mechanism for protocol upgrade that it brought to the table should have been standardized separately in order to make it so when http&#x2F;2 was actually ready, we&#x27;d be able to move to it.
dcsommer超过 10 年前
I&#x27;m wondering what can be gained by discussing this exact email again? See <a href="https://news.ycombinator.com/item?id=7798946" rel="nofollow">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=7798946</a><p>We&#x27;ve also had more recent PHK rants on HN since then, so I don&#x27;t see what we gain from a plain repost, with no additional context.
评论 #8882823 未加载
kemayo超过 10 年前
It does seem like if they are already saying that an HTTP&#x2F;3 is going to be needed then they might as well not bother releasing HTTP&#x2F;2. Implementors are going to have minimal motivation to actually implement it, and will probably hang around waiting for the improved version that they know is coming.
评论 #8882915 未加载
debacle超过 10 年前
On the one hand, any HTTP 2.0 spec that still has cookies is a travesty.<p>On the other hand, I think we need to get some input from dissenters other than PHK (not that I don&#x27;t agree with him).
评论 #8882912 未加载
评论 #8882885 未加载