Thursday, 25 March 2010

Fun With BCC

Luis Villa falls into the old "reply to all on a BCCed email" trap...

I think it's safe to tell a little story about the best case I was involved with. When I was finishing up my PhD in 2000, Jim Larus at Microsoft Research asked me to come up for a visit so they could encourage me to apply for a job there. I declined, carefully and politely explaining that I'd already decided I did not want to work at Microsoft (because of the issues around the company as a whole; MSR was fantastic and I liked the people there). I BCC'ed my PhD advisors on this reply because they needed to be aware of the situation. One of my advisors replied "Stick it to 'em!" ... to all, of course. Hilarity all around.

BCC is a dangerous and obsolete tool IMHO. It's safer to just forward your message to the secret recipients.

Amor Robustum

I suppose I'd better give my take on John Gruber's spiel, especially because Mark Pilgrim was pushing a similar line on #whatwg:
But if Mozilla’s position were really about idealism — tough love for the good of the web in the name of free, open file formats — then in addition to not supporting H.264, they’d drop support for plugins like Flash Player.

Let's assume for the sake of this argument that eliminating NPAPI plugins (and their shapeshifted ActiveX cousins) would be a good thing for the open Web. Would dropping support for NPAPI plugins in Firefox be a good thing for the open Web? Not necessarily. If we dropped support for Flash in desktop Firefox, then (since desktop browsers are so competitive and the switching costs are quite low) a very large number of users would stop using Firefox and start using another browser that does support Flash. So Flash usage would not decrease much. We would certainly not even get close to eliminating the use of Flash on Web pages. On the other hand, there would be significantly less competition in the browser market. Overall, that's not a win for the open Web. Pointless martyrdom can be fun, but it's not good strategy.

The situation with the HTML5 <video> element and H.264 is very different. At the moment, lack of support for H.264 in <video> is not driving users away from Firefox. That is a battle we can still fight and possibly win without destroying our relevance.

Tuesday, 9 March 2010

Rainbows End

I just finished reading Vernor Vinge's Rainbows End. Pretty good, but I think A Fire Upon The Deep and A Deepness In The Sky were better. I think in Rainbows End the technology gets a bit in the way of the story ... Vinge carefully prognosticates and extrapolates, and that makes it overwrought.

Having said that, the prognostication is quite good and worth reading and thinking about. And having said THAT, there are a few major blind spots. In Vinge's world, software usually works, porn and spam are almost absent, and nations with extraordinary totalitarian power don't abuse it. Hmm ... I doubt it :-).

Thursday, 4 March 2010

Nightmare On Infinite Loop

This Apple patent attack on HTC is very bad news. Like Microsoft's attack on Tomtom, it's a offensive lawsuit based largely on software patents. Unlike Microsoft's attack, it seems to be not about licensing deals but about shutting down the competition, and many of the patents involved are for obvious, fundamental and very general software techniques. LWN has a good article with links to the specific patents involved.

Consider the first claim of patent 7,362,331:

1. A method for moving an object in a graphical user interface, comprising the steps of:
a) determining a path of movement for the object along at least one axis, and a period of time for the movement along said path;
b) establishing a non-constant velocity function along said axis for said period of time;
c) calculating an instantaneous position for the object along said path in accordance with said function and the relationship of a current time value to said period of time;
d) displaying said object at said calculated position; and
e) iteratively repeating steps (c) and (d) during said period of time.

It was filed in 2001 and issued in 2008. (The other claims are mostly restatements of the same idea.) So when Jobs says
"We think competition is healthy, but competitors should create their own original technology, not steal ours."

not only does he believe that in 2000 Apple was the first company to invent the idea of time-based animation of objects with non-constant velocity, but he also believes that they deserve a monopoly on use of that idea until 2021, and that anyone else using the idea until then is stealing from Apple. All three beliefs are ludicrous and shameful.

It's worth noting that any implementation of CSS transitions would infringe patent 7,362,331. I hope Apple isn't planning to sue implementers of CSS transitions for "stealing" their "technology".

That patent was just the first one in LWN's list. Some of the others look worse.

I'm very glad I don't work for Apple.