...looks like they granted this one as well

http://www.patentstorm.us/patents/7346672/description.html

now I know full well there is prior on this!

I'd like to patent the system whereby is a human is born they then proceed to old age.

I'll only charge 50p per lifetime use.

Mmm, £3B for me..

Is it just me that thinks the whole patent business has gone barking mad in the past few years?

Surely the principle of auto updates shouldn't be patentable. Maybe a specific technical solution to the problem, but the principle in the frist place? Madness. Have Apple patented the on/off switch as well? How about 'the chair' - a means of plonking your backside on something whilst you use our patented on/off switch, so as to relieve the strain on your legs.

Or, if they want to sue a bunch of other manufacturers, the legtop computer: a small, portable computer that can be placed on the top surface of your legs whilst in a seated position. Then they can drag all the 'laptop' manufacturers through the courts for IP infringement.

The patenting of sections of the human genome was the start of it this bonkers practice - where will it end?

/rant_off

I know that's the case for Copyright and Trademark, but IANAL.

Internet updaters have been a ubiquitous fact of ONline life for several years now. Why is this thig suddenly lurching out of the grave, after well over a decade?

One wonders why they didn't ask Apple for a free license first..

OTOH, the time that standards commitees take, the patent will have expired by then. Isn't it 18 years?

I'd call auto-update way too obvious to patent. Maybe it could be overturned on those grounds.

Alternatively, of course, the W3C could take out a patent on auto-updating using signatures and PKI to guarantee integrity and prevent spoofing or MITM attacks, which is how all modern updaters work, and then tell Apple they're welcome to their lousy years-out-of-date insecure patented method.

it describe a process that is obvious, infact I am quite sure that none of the people who added the auto-update feature in their application have ever heard of this patent.

I understand that Apple wants to protect its patent, but considering how common the use of auto-update is, why are they trying to block it from becoming a standard? they haven't blocked anyone else from using it! (as far as I know)

They might not be able to claim prior art (there may or may not be earlier systems doing this) but certainly it fails the obviousness test?

Any technically minded person faced with a problem of keeping a group of networked machines updated witha piece of software would undoubtedly come up with a solution that covers these broad strokes.

Also, if they believe there's any validity in the patent why haven't they sued everyone already for breach? Or have they already got licensing agreements with every company that does this?

Surely not-prior-but-art-thought-up-without-any-knowledge-of-the-original-patent-and-is-now-used-everywhere should be enough to invalidate the patent?

It should be the company's job to enforce the patent *before* everyone starts using it rather than pulling it out at a later stage..

Oh.. but that would be sensible wouldn't it?

I remember Apple Link Personal Edition back in 1989 or so updating itself when there was a new version available. Around the same time I set up software build systems that would look for updates to code and refresh your local copy before compiling it and running it. It was an obvious thing to do.

Almost as useless as SCO stock....

Why do we still have such a flawed system. I uinderstand that companies don't want someone to steal their products, rebrand it and sell it, but this is rediculous.

html works for me as a scripted language

and httpd code 200 tells me to get a newer version of the script i want to run

The world's first Internet worm, Morris, sort of auto updated by mistake owing to a bug in the code in November 1988

See http://en.wikipedia.org/wiki/Morris_worm

I was contracted to a large utility to write and run a dBase IV system which did this in 1993: sales reps using laptops would have new software installed when they either dialled in or plugged into the office network & server.

I can supply more details if anyone's interested.

Paris 'cos her software was under control back then too.

In the 1980s the Acorn Archimedes would advise you that various parts of the OS were out of date and invite you to update them (when installing new software for example). Changing this to actually doing the upgrade for you was trivially obvious, once there was a method of actually doing that!

"[I]t is desirable to provide a mechanism by which software programs can be automatically updated in an easy and effective manner without concerns over security issues that are raised by external access requirements. ... In accordance with the present invention, this objective is achieved by means of a method and apparatus that permits a software program running on a computer to automatically replace itself with a newer version in a completely automated fashion, without interruption of its primary function, and in a manner that is completely transparent to the user of the computer."

Notably,

- The process is completely automatic, with no interaction or notification to the user.

- The entire program itself is being replaced, not [just] data files such as virus signatures.

- Everything is done by the program itself, not a separate updater, background service, periodic job, nor does it hand off responsibility to another program on an update server.

@Julian I-Do-Stuff:

Yes, if server-side code determines if a newer version exists, it is not covered.

@AC 15:58:

If-modified-since doesn't count because it is at most a component of the update process (it won't move and replace the currently-running program), and the evaluation is done server-side. Also, using that header for program version checking is an extremely non-obvious use of the protocol.

@Simon Booth:

The Morris worm was willing to infect a computer multiple times, which has nothing to do with updates. As far as I know, there was only one version of it.

@Owen Smith:

No, code 200 just indicates that the request was successful. Unless you're making a completely bizarre use of HTTP, with, say, code 300 indicating there is no newer version, which passes the obviousness test but fails the sanity one. And it does the version checking server-side.

so many times on The Reg we get the foaming at the mouth and the rants from people who haven't read *and understood* the point of a software patent and then drag in the "oh, I shall patent oxygen and charge you all for breathing" bit ...

... Yes, the essential bit that I needed to see pointed out is that it is a program replacing itself with a newer version WHILE CONTINUING TO RUN. Not Firefox saying "there's a new version, do you want to download it and install it before Firefox starts, or do you want to download it now, and it will be available next time you start Firefox" but actually Firefox continuing to work but it's magically updated from 3.0.10 to 3.0.11 without interupting your browsing.

I can't think of many/any applications that do that now ... and I certainly don't know of any before 1995.

Even now most applications seem to have the "a new version has been downloaded, you will need to restart Windows to complete the installation".

Even anti-virus software usually requires you to start scanning again if you download a new virus definition file, never mind a new version of the running software.

Nevertheless, I'm not sure whether it is an original enough idea to warrant surviving a patent challenge.

And to those going FFS this should never have been patented ... the current process seems to be that unless it's really obvious or there is obvious prior art, then the patent is granted and it is left to other companies to challenge it. It makes it easier/quicker/cyheaper to apply for a patent and to be granted a patent, and it is left to the community to see if anyone cares.

What you are describing is that patents are just simply based on the first person to register the idea, regardless of it being so blindingly obvious. It's like saying the first man who made fire could therefore patent it, or even the first man who learned to walk!

That's not the spirit of patents, especially where a huge corporation is using it to subsequently sue every man and his dog knowing full well they are almost certainly going to come across the idea.

There's also a part of the patent spirit which says you must actually enforce your patent. Just sitting on it for years until someone comes along with the same idea so you can sue them is utterly wrong.

As for "there are no self updating without interruption" apps out there. How about Google Chrome? It's been updating me for ages now and I've never noticed. Okay it wasn't developed before 1995 ;). Still, I'm sure there are plenty of stuff from the old client/server days of the 80s and early 90s that featured this kind of update, especially as uptime for business systems was a critical factor.

As for Apple. Does their patent require that along with the update, other Apple branded applications must be forced upon the user at the same time (iTunes & QuickTime for example)? ;)

Hi all,

I'm W3C's PAG chair and I have read many comments with a lot of interest. Some comments where posted anonymously without URI to the reported prior art. This makes it very hard for me to get to that useful information. I would invite those commenting on Plan9 and TC 500 to send me private email with a hint where I can find the information given (Maybe I have to go to the library). Or just put a URI here. And please, if you have any further idea, do not hesitate to send comments here, to my personal email address or to the public mailing-list:

personal: rigo@w3.org

list: public-widgets-pag@w3.org

Thanks for contributing to the Web's well being

Rigo Wenning

W3C Legal counsel

Things which are perfectly obvious should not be patent-able.

This is patents stifling innovation again. And Apple is not the only offender here....