Software patents part 2

Picking up where we left off yesterday. Only two more patents to go.




US 6,339,780. Granted patent published 15 January 2002. Filed 6 May 1997.

Claim 37 (Claims 1-36 presumably deleted during examination)

1. A hypermedia browser embodied on a computer-readable medium for execution on an information processing device having a limited display area, wherein the hypermedia browser has a content viewing area for viewing content and is configured to display a temporary graphic element over the content viewing area during times when the browser is loading content, wherein the temporary graphic element is positioned over the content viewing area to obstruct only part of the content in the content viewing area, wherein the temporary graphic element is not content and wherein content comprises data for presentation which is from a source external to the browser.

Another potted summary of the patent, although in this case the actual patent is fairly short. Web browsers at that time often included a little animated icon to indicate that a web page was being downloaded. On a small screen, this is a waste of screen space. Solution, draw animated icon over the main viewing area of the browser while the page is loading, then get rid of it once the page has loaded.

My thoughts on this are remarkably similar to my thoughts on US 5,778,372 aka the 'prioritise important stuff' first patent.

US 6,891,551. Granted patent published 25 July 2002. Earliest filing date 10 November 2000

Claim 1.

1. A method for selecting portions of electronic data on a display device, comprising the steps of:

generating a selection area identifying a first portion of said electronic data, wherein said selection area includes one or more selection handles;

receiving an input from a user associated with said one or more selection handles; and

resizing said selection area responsive to said user input.

OK, this is quite an interesting one and looks rather similar to the method used for selecting text in recent versions of iOS. Indeed the rationale behind this invention was very similar. Clicking and dragging to select text on a computer screen isn't particularly accurate, so lets add resizing handles so that the user can tweak an existing selection box, rather than starting over from scratch if the wrong text has been selected.

Without knowing much about the state of the art back in 2000, I'm not really sure how novel this is. Certainly, the only similar example of a similar system, I can think of is in the far more recent iOS. With regard to inventiveness, my first thought was 'well if you want to resize a selection box, how else would you do it.' Then after a couple more minutes thought, it was clear that there were a number of other ways. For example you could use keyboard cursor keys to manipulate selection box instead of using the resizing handles, or use alternative mouse controls, e.g shift-clicking, to move the boundaries of the selection box.

As I commented in my previous post on the Apple slide-to-unlock patent, the presence of alternatives doesn't necessarily mean that your chosen option isn't obvious but it does make it harder to argue that it is obvious without some kind of steer from the prior art. Also, like the slide-to-unlock patent, I find it difficult to get too worked up about this one. The patent describes a useful invention (arguably more so than slide-to-unlock), which is narrowly claimed and fairly straightforward for competitors to invent around.


So where does that leave us. One patent that I don't honestly feel I can offer an opinion on and two which are - not exactly poster children for software patents as a driver of innovation. One more which I'd quibble about but which I'm prepared to concede does disclose a useful invention. Finally one which looks like it should be lumped in with the 'poster children' at first glance but which doesn't look too bad when considered in a bit more detail.


All in all, not a particularly inspiring bunch but one which collectively doesnt quite deserve the scorn heaped upon it by assorted internet commentators.

Software patents - don't dismiss them.

This post was written as a follow up to a comment I made on the Ars Technica forums. One contributor made some (fairly predictable) comments about the validity of a set of Microsoft patents on the basis of the patent abstracts. I pointed out that abstracts weren't the best way of doing this and that to properly judge a patent you need to read the claims. I then commented that it is quite common for overly broad and silly claims to get whittled down into something sensible during examination.

Naturally I got curious at this point. Certainly, these particular patents didn't look especially earth shaking from their abstracts but did the claimed inventions bear any resemblance to those abstracts. Unfortunately for me, they did in this case, although with a bit more thought, my initial knee-jerk dismissal of them was a little unfair.

This post also turned out to be rather lengthy, so I'll split it into two and post the second half tomorrow. With that said - lets have a look at those patents.

US 6,957,233. 

Granted patent published 18 October 2005, filed 7 December 1999.

Claim 1.

A computer-implemented method for annotating a system having a display for displaying a page having user selectable objects being intermixed with markup tags, said user selectable objects and said markup tags being stored in a non-modifiable portion of a file, said mark-up tags bounding said user selectable objects in said file, comprising the steps of:

receiving user input for a selection of an object of said user selectable objects on the displayed page;

receiving user input for providing an annotation associated with said selected object; said annotation being modifiable;

determining a position of the selected object in the non-modifiable portion of the file regardless of said markup tags bounding said selected object;

storing the position and the annotation separately from the non-modifiable portion of the file; and

providing a portion of said display configured for navigating to the previously selected object based on said position, when said annotation is subsequently selected.

OK this is a little bit opaque, so a bit of context from the patent description would probably help. This patent was all to do with electronic documents and making them more akin to actual paper documents by providing a way of annotating them. I admit to being slightly bemused by this because the last thing I do with a nice shiny book is scribble notes in it but each to their own. Anyway, it turns out that the current method of doing this at the time was to edit the document file to include said annotations. Apparently this presented a problem in that some copyright holders weren't too happy about electronic editing of copies of their work. Hence there was a need for a way of annotating electronic documents without editing the actual document file.

In a nutshell that's what this patent claim is all about. Find out where the user wants to place an annotation, analyse the document file to find the location of that annotation and then store the annotation and its associated position in a separate file.

Back of the envelope commentary

The key part of the claim is that location finding step. Even with a cautionary mental note that most things are obvious in hindsight, if you want to annotate a file without editing it to include the annotation, then it seems to me that the obvious solution is to store that annotation in a separate file. However, I can well imagine that there could be an invention in finding a suitable way of mapping the annotation file onto the document file. Without knowing anything about the prior art, I'll give this claim the benefit of the doubt as regards novelty and inventive step.

However, the description only discloses a limited number of ways of achieving the mapping step and appears to be primarily limited to text files. Under US patent law, a patent description must be enabling and should be commensurate in scope with the claims. In other words, as per the Federal Circuit “the specification must teach those skilled in the art how to make and use the full scope of the claimed invention without 'undue experimentation'.” On that basis, I would argue that this claim is too broad.

US 5,778,372. 

Granted patent published 07 July 1998. Filed 18 April 1996.

Claim 1.

1. A method of remotely browsing an electronic document residing at a remote site on a computer network and specifying a background image which is to be displayed with the electronic document superimposed thereon comprising in response to a user's request to browse to the electronic document:

requesting the electronic document from the remote site on the computer network;

receiving the electronic document from the remote site;

requesting the background image specified in the electronic document from the remote site on the computer network;

receiving the background image from the remote site;

drawing an initial display of the electronic document without the background image prior to receiving the background image from the remote site; and

redrawing the electronic document superimposed over the background image after receiving the background image from the remote site;

whereby the initial display of the electronic document is not delayed until the background image is received from the remote site.

A very condensed summary of the patent description. Web pages with images can be slow to load. Slowly loading webpages

Seriously? I appreciate that this was all pretty new stuff back in 1996 and so there most likely wasn't much prior art around but honestly. Prioritising bits of a webpage to download, so that quick to load, important stuff is downloaded before long to load, not so important stuff seems like a very obvious solution to the described problem.

I really feel for the examiner who got this one. This is an application that cries out to be rejected 'just because' but as I've mentioned previously that doesn't tend to work very well.

US 5,889,522. 

Granted patent published 30 March 1999. Filed 13 December 1994.

Claim 1

1. In a computer system having a display, an input device, and a processor running an operating system (OS) and an application program, the application program running in an application window having a client area, the client area for displaying and manipulating application data, a method comprising:

(a) outputting a control window to the display, the control window being associated with the application program;

(b) predefining, by the OS, a tab control class for providing information to the application program, wherein:

(i) the tab control class includes a plurality of tab control images, each tab control image defining a page having a tab; and

(ii) each page displays application parameters from the application program;

(c) instantiating, by the application program, the tab control class; and

(d) outputting, to the display, a first tab control image superimposed on top of other tab control images within the window, such that the first tab control image, any application parameters thereof, and the tabs of the other tab control images are viewable in the control window.

OK, to me this looks like a claim to a tabbed preference panel, familiar to any Windows (or Mac OS X for that matter) user. Single window with all the preferences for a particular program, organised in groups, each group on its own tab, click the tab to bring up a given group.

I have to confess that I have no idea about this one. The patent was filed in 1994, so this is pre Windows 95 we're talking about. Nope, I'm simply not familiar enough with the art to make a sensible comment on the likely novelty or inventiveness of this one.

Unlocking Innovation

As reported by the IPKat, Apple have been awarded US patent 8,046,721 for 'unlocking a device by performing a gesture on an unlock image'. This is a continuation of US 7,657,849 which was recently rejected by the Dutch courts for being obvious. A key piece of evidence used by the Dutch courts was the Neonode N1m phone which, as shown in this video, can be unlocked by swiping a finger from left to right across the touch screen.

I thought that this was an interesting case for a number of reasons. Firstly, it's almost a textbook example of a 'patent likely to generate ridicule by the general public.' Is this justified? Secondly, if US 7,657,849 was rejected, is US 8,046,721 likely to fare any better? Last but not least, regardless of the merit of the patent it is an extremely narrow patent for a very specific invention. With that said, does it actually matter to anybody other than Apple that this patent was granted?

By way of a quick recap, in order to be granted, a patent has to be (amongst other criteria) both novel and inventive. In other words, the invention disclosed in the patent can't have been previously made available to the public but neither should it be merely a trivial variation on a previously disclosed invention.

Novelty is reasonably straightforward to decide. Check the patent claim, pull out all the features recited in that claim and compare with the previous disclosure (referred to as prior art). If the prior art contains all the features of the claim then that claim lacks novelty. In practice, its a little more complicated than that (and you'd struggle to find a piece of case law that deals with novelty in quite that way) but I think that captures the essence of it.

Inventive step or obviousness on the other hand is rather different and can be extremely difficult to decide. If the patent claim includes features that are not described in the prior art, then the claim will be novel. The question then is whether those extra features add anything to the claim, or whether they're relatively minor things that any reasonably skilled person would come up with as a matter of course after reading or being shown the prior art. Similarly, if there is no prior art at all, then the claim is certainly novel but is not necessarily inventive – it may just be a collection of trivial features that nobody has bothered to try and patent before. However, it will probably be a lot harder to argue that an invention is obvious, if nothing even vaguely like it has been made available to the public before.

The tricky thing about inventive step, is that most things are obvious in hindsight. To properly decide whether something is obvious or not compared one or more pieces of prior art, you need to consider what would be obvious to somebody reading that prior art at the time at the time the patent was filed. This isn't easy to do because it's a necessarily artificial point of view to have to take. Equally tricky, from the point of view of invalidating a patent, can be constructing a reasonable argument that will stand up in court. Arguing that something is obvious because 'look at it - it just is' rarely works. There has to be a valid reason why.

For the Apple 'slide to unlock' patent, in my opinion (for whatever that's worth), the Dutch court got US 7,657,849 right. Claim 1 of this patent reads as follows:

1. A method of controlling an electronic device with a touch-sensitive display, comprising: detecting contact with the touch-sensitive display while the device is in a user-interface lock state; moving an unlock image along a predefined displayed path on the touch-sensitive display in accordance with the contact, wherein the unlock image is a graphical, interactive user-interface object with which a user interacts in order to unlock the device; transitioning the device to a user-interface unlock state if the detected contact corresponds to a predefined gesture; and maintaining the device in the user-interface lock state if the detected contact does not correspond to the predefined gesture.

I strongly suspect that the same Dutch court would also reject US 8,046,721. Claim 1 of this patent reads as follows:

1. A method of unlocking a hand-held electronic device, the device including a touch-sensitive display, the method comprising: detecting a contact with the touch-sensitive display at a first predefined location corresponding to an unlock image; continuously moving the unlock image on the touch-sensitive display in accordance with movement of the contact while continuous contact with the touch screen is maintained, wherein the unlock image is a graphical, interactive user-interface object with which a user interacts in order to unlock the device; and unlocking the hand-held electronic device if the moving the unlock image on the touch-sensitive display results in movement of the unlock image from the first predefined location to a predefined unlock region on the touch-sensitive display.

This is a little bit more specific than the previous claim (and corresponds more directly to the actual iPhone unlock screen) , in that the user has to maintain continuous contact with the screen during the unlock gesture and has to explicitly touch the unlock image to interact with it.

However, in both cases, the only significant difference between the claim and the 'swipe to unlock' gesture used on the Neonode N1m, is the addition of a user interface widget to provide a visual clue to the user that a particular gesture will unlock the phone. Adding such a visual clue certainly makes the phone easier to use but I don't think that adding such a clue is inventive. Arguably, that is a major point of any graphical user interface – to provide visual clues that interacting with the interface in a particular way will have a desired effect on the device running that interface.

On the other hand, without the Neonode prior art it is a lot harder to see how you would set about arguing that either claim is obvious. The same argument about providing a visual clue would still apply but why have that particular gesture to unlock the phone? For that matter, why use the touchscreen at all? Why not just have a physical unlock button? If having a minimal number of physical buttons is important to your phone design, why not press the physical button once to activate the touchscreen and then again to unlock the phone? Or hold down the physical button and perform a gesture with the whole handset, e.g. a flick or a shake (assuming your phone includes an accelerometer). For that matter, if you have your heart set on using the touchscreen to unlock, why not simply designate a number of points on the screen that have to be touched simultaneously to unlock the phone? Both the slide-to-unlock patents specify a moving unlock image, so using a non-moving unlock image avoids both patents.

Having a number of alternative options isn't necessarily enough to make any one of those options non-obvious. However, by the same token, if even a non-skilled person can think up four alternative options for unlocking a phone, then I think it also becomes more difficult to argue that a different option, absent any clue from the prior art, is obvious.

For the same reason, for the hypothetical case, where there wasn't any prior art for unlocking a phone with a touchscreen gesture, I personally wouldn't have any problems with either of these slide to unlock patents. They're far from earth shaking but they do solve a problem in a distinctive and eye catching way, so I can understand why a company would want to try and protect that particular solution to the problem. Yes, they do prevent competing companies from using that solution but neither is it particularly difficult to find alternative solutions that avoid the patents altogether. Slide-to-unlock is very much a nice to have feature than one that would be a serious detriment (or indeed any kind of detriment) to competitors if they couldn't use it. In which case, I would find it hard to really get too indignant about the patents.