Archive 14. November 2009

Fefe's Criticism of SPDY - I'm not exactly a die-hard Fefe fan, but his criticism of Google's new protocol is, in my opinion, quite valid. My opinion on SPDY is also rather negative - the points of criticism that Google has with HTTP could be easily resolved within HTTP. Multi-Request? We already have keep-alive to minimize connection setups, expanding this to multi-request where you send multiple requests at once and then immediately get all the data would not be an unsolvable problem. SSL has been available for HTTP for a long time. Compressed headers? Sorry, but the headers really don't make up the large part of the data, compressing them doesn't really bring anything.

More Freedom Necessary as Top Developers Abandon iPhone - if Apple doesn't get this under control soon, it could have a pretty negative impact. Because if die-hard Apple shops like Rogue Amoeba are already showing Apple the red card, others will follow in the long run. And if only silly flashlights are left burning in the App Store, Apple might realize that their approach was rarely stupid. But then it will be too late ... (just look at how Palm more or less led the Palm Pre platform ad absurdum through all the fuss around their variant of the App Store).

nothing new - someone compares Go (Google's new system language) with Algol 68 - and the old lady Algol 68 comes out quite well.

Why Common Lisp will never really become mainstream - the linked source is only used to use a binary-ascii decoding/encoding library in various Common Lisps via an automatically decoding and encoding stream. What's inside? Mountains of #+ markers with various Common Lisp implementations. That's not portability, that's just a mess.