Comment by Animats | Hacker News Reader

Comment by Animats | Hacker News Reader

The link to the patent 7,517,222 is wrong; it's to an application with the same title. Here's the key claim:

1. A magnetic connector system comprising: a first connector having a first plurality of electrical contacts and a plurality of magnets; and a second connector having a second plurality of electrical contacts and a magnetic element, the second plurality of electrical contacts being adapted to mate with the first plurality of electrical contacts when the first connector couples to the second connector, wherein the plurality of magnets of the first connector are proximally located and are arranged in opposing polarities with respect to each other so that when the first connector is brought in close proximity to the second connector, magnetic field lines travel through the magnetic element of the second connector from one of the plurality of magnets in the first connector to another one of the plurality of magnets in the first connector, thereby increasing magnetic attraction between the first connector and the second connector.

All the magnets are in the laptop end of the connector pair, not the cable end. Selling only the non-magnet end might not be infringing, since that's just a metal plate around some contacts. You might be able to beat this. But it would be tough.

If you really wanted to pull Apple's chain, you could make a better connector than Apple makes. You could fix the problem with their smaller magnetic connector, which is known to fall off too easily. Here's how.

There's an exotic device called the Correlated Magnetics MagPrinter (http://www.correlatedmagnetics.com). This is a CNC machine which magnetizes magnets. Not just ordinary uniform magnets, but single pieces of material with custom-designed magnetic patterns. With this, you can make one piece of magnetic material behave as if it is multiple magnets. You can create magnets with various self-aligning properties. You could probably make a plug for Apple's socket which didn't look like Apple's, didn't have a "plurality of magnets", mated with their connector, and worked better.

TechShop in San Francisco has a MagPrinter, and any member can use it. They also have the software you need to design the magnetic fields and create the files for the MagPrinter.

So get busy and out-invent Apple.

Animats, your idea is great and it's something we considered a lot when we were thinking about the connector. We almost went and did it. However there were a couple of problems with that patents (as a side note - my co-founder is actually a former patent troll, so we dug quite deeply into the patents). Our priority was to not infringe on any of Apple's patents, so we were as thorough as possible with our research.

Here are the main patents around the MagSafe: http://www.google.com/patents/US7645143 http://www.google.com/patents/US7901216 http://www.google.com/patents/US8087939 http://www.google.com/patents/US8177560 http://www.google.com/patents/US7311526 (the key patent)

Here are the main claims (apart from the one you mentioned): (a) A first connector comprising: a first plurality of electrical contacts, the first plurality of electrical contacts to mate with a second plurality of electrical contacts when the first connector couples to a second connector, wherein the first plurality of electrical contacts consisting of a central contact to convey a signal, two contacts to convey a power supply, one contact on each side of the central contact, and two contacts to provide a return path, one contact on each side of the central contact, wherein when the first connector couples to the second connector, the first and second plurality of electrical contacts define a corresponding plurality of electrical paths; and a magnetic element, the magnetic element to mate with a plurality of magnets in the second connector that are proximally located and arranged in opposing polarities with respect to each other so that when the first connector is brought in close proximity to the second connector, magnetic field lines travel through the magnetic element of the first connector from one of the plurality of magnets in the second connector to another one of the plurality of magnets in the second connector.

(b - part of it) a plurality of movable first contacts to make electrically conductive paths with a plurality of second contacts in a second connector when the first connector is mated with the second connector, each of the movable first contacts biased by one of a plurality of first springs;

The biggest problem is with the claim that speaks about a magnetic field from multiple magnets (4 are present on computer side) traveling through a magnetic element (ie. metal) on the connector side, which the patent states creates a stronger attraction. We thought it would be possible to place metal bits on the connector side in such a way that the magnetic field would only be passing through one element - thus creating less of a magnetic attraction but being a different invention. Unfortunately, at this point it becomes a matter of proving wether or not that's what's actually happening (this was our biggest worry as burden of proof would be on us).

There were another 2 - 4 other claims that also made it hard to improve upon (also not infringe on any of their patents) and create our own connector - not to mention the design patents.

I was actually thinking of writing a blog post about this issue, it was extremely interesting to look into all of this.

The problem isn't defeating the patent technically - it's defeating it legally, while still remaining profitable. And nobody has more money to throw at lawyers than Apple, Inc.

Reply on news.ycombinator.com