Live data from Hacker News

Hello HTTP/2, Goodbye SPDY

blog.chromium.org

11–20 of 181 posts

Re: Hello HTTP/2, Goodbye SPDY

#11
post #6

What happens if someone built a service based on it? Should they never trust browsers keeping alive even the shitty (in comparison to free and standardised HTTP/2) features? What's great about the web is that now 20 year old services still are working in the latest runtimes (browsers).

Nothing is being lost with regards to SPDY. Old versions were never supported; if you wanted to use SPDY you had to commit to keeping the server updated.

So effectively they have just announced a long term support edition of SPDY. What an odd time to complain about the lack of long term support.

Re: Hello HTTP/2, Goodbye SPDY

#12
post #8
post #6

What happens if someone built a service based on it? Should they never trust browsers keeping alive even the shitty (in comparison to free and standardised HTTP/2) features? What's great about the web is that now 20 year old services still are working in the latest runtimes (browsers).

It is always risky to build a service based on something that is not yet standardized. SPDY was in progress to be standardized, but the process ended up with parts of it in HTTP2, making SPDY unnecessary, as I understand things. It would be the right thing for Google to remove SPDY at this point, otherwise it would be running a nonstandard protocol that other browsers do not, which can lead to fragmentation - as we s…

> It would be the right thing for Google to remove SPDY at this point

It's going away, just maybe not soon enough for everyone's tastes. From the blog:

> We plan to remove support for SPDY in early 2016

Re: Hello HTTP/2, Goodbye SPDY

#14

SPDY came and went before I had to implement it. Phew. On a serious note: it's nice to see ALNP being used in HTTP/2

ALNP has been used with SPDY for a while now. It one of the nice improvements that fell out of testing/iterating SPDY in public. The NPN approach was a bad idea since the client drove what got picked (With NPN, Server tells the client what other protocols it supports in the ServerHello and the client picks whatever it wants. ALNP reverses that.)

Re: Hello HTTP/2, Goodbye SPDY

#16
post #8

Earlier quoted context omitted.

It is always risky to build a service based on something that is not yet standardized. SPDY was in progress to be standardized, but the process ended up with parts of it in HTTP2, making SPDY unnecessary, as I understand things. It would be the right thing for Google to remove SPDY at this point, otherwise it would be running a nonstandard protocol that other browsers do not, which can lead to fragmentation - as we s…

> It would be the right thing for Google to remove SPDY at this point It's going away, just maybe not soon enough for everyone's tastes. From the blog: > We plan to remove support for SPDY in early 2016

I think that's a reasonable timeline, actually - sorry if what I wrote was confusing to imply "right now". I meant to say "It would be the right thing for Google to announce the timeline to remove SPDY at this point in time."

A year head's up gives people plenty of time to update their sites, and sounds fair and reasonable.

Re: Hello HTTP/2, Goodbye SPDY

#17

Looking forward to when HAProxy support for HTTP/2 lands since they refused to implement SPDY support. Here's a list of common servers support for SPDY/HTTP2: https://istlsfastyet.com/#server-performance

Current HAProxy already supports the handshake of SPDY/HTTP2 via NPN and ALPN. You have to route to proper backends. You also need to provide a HTTP/1.1 fallback implementation for incapable clients. Once setup that works very well. I am using it for our blog (https://blog.cloudno.de)

Re: Hello HTTP/2, Goodbye SPDY

#18
post #15

Is HTTPS mandatory on HTTP/2 like it was on SPDY?

> Is HTTPS mandatory on HTTP/2 like it was on SPDY?

Not in terms of the protocol spec, but most major browser vendors have indicated that they only intend to support HTTP/2 in-browser over TLS connections, so in practice for typical, browser-targeting use cases, it looks like it will, at least initially.

Re: Hello HTTP/2, Goodbye SPDY

#19
post #8

Earlier quoted context omitted.

It is always risky to build a service based on something that is not yet standardized. SPDY was in progress to be standardized, but the process ended up with parts of it in HTTP2, making SPDY unnecessary, as I understand things. It would be the right thing for Google to remove SPDY at this point, otherwise it would be running a nonstandard protocol that other browsers do not, which can lead to fragmentation - as we s…

To be fair Google did happily kill Gears when HTML5 became a viable [early draft] standard.

Agreed, Google did the right thing to remove Gears.

My concern is because, overall, Google has a bad track record in this area: FileSystem is still enabled, WebSQL is still enabled, PNaCl is still enabled edit: and H.264 was never removed despite announcing the intent to do so.

Post reply on HN