Hacker News new | past | comments | ask | show | jobs | submit login
Rivals can create copycat software through testing and user manuals: UK court (out-law.com)
77 points by epo on Nov 29, 2013 | hide | past | favorite | 30 comments



Heh. Funny story - I was once asked to clone the Foxtel STB in Australia for a competitor - they wanted to be able to take the box along to a demo at Foxtel, plug it in, and show the same interface as the real Foxtel STB, but better. So, taking the company's in-house middleware, and a UI framework that I had recently written, I went about cloning the Foxtel box, screen by screen, keypress by keypress.

This was in the 90s, and STB interfaces were in their infancy - there were no animations, or alpha blending, or huge databases of films, including images of jackets etc - it was mostly just text and some simple vector graphics. All of which made my life as a programmer much easier, thankfully, because I only had about a week to get the thing done.

The funny part of the story comes in when I discovered that the software was just full of bugs. I would press a button on the remote control on a certain screen and the STB would reboot. This proved to be a bit of a problem for me, because I had to try and figure out what we were actually supposed to do, but as I wasn't allowed to disassemble the actual Foxtel code and see what it was trying to do, I was quite perplexed.

3 days into the week my boss was starting to get edgy - I had all of the most common screens up and running, but he could see that I had logged a whole bunch of bugs into the bug tracker, and I apparently wasn't making much progress in getting them fixed. So he came by my desk to see what was going on. I explained to him that the bugs existed in the original STB that I was copying, and the best replacement functionalities that I could come up with led to quite a bit of complexity in the code base. My boss looked at me quizzically, and then just shrugged his shoulders "Just dereference a NULL-pointer!".

Oh. Right. We never did get a contract from Foxtel, but I gave one of the cloned STBs to a neighbour to test in real-life conditions. I couldn't pry it out of his hands afterwards though - he thought it was so much better than the original product, dereference NULL pointers and all...


For those like me who didn't quite understand, STB stands for set - top box.

Also, that's a pretty cool story actually!


So if you make something that so much as uses a shopping cart you violate 394 patents, but if you grab the manual and make something virtually identical you're in violation of no copyright?

I get that patents != copyright, but we're still entering a rather strange state of affairs...


As a UK/EU citizen, this has always been my understanding of the law, and it's not strange to me.

I might violate US patent law (which is an increasing concern given the apparent freedom of extradition), but I'm not violating UK law.

There is copyright over my writing (here, lines of code), but typically not over my idea.


No software patents in Europe (well, not officially anyway).


Yeah, but european investors ask for patents right off the bat...


Not just investors. Anyone I've met that even mentors startups or works with them through government programmes here pushes for patents on everything. It's annoying to see for someone aware of how badly patents and patent trolls are screwing up the software industry.


This is in the UK, not Texas.

AFAIK, software patent trolling isn't very bad in the UK


Very bad? I thought we didn't recognise software patents how could it even be possible? US Import law maybe?


It's always been the case that you can accidentally violate patents but generally can't accidentally violate a copyright. That isn't a state we're entering — it's inherent in the two concepts.


Actually them being separate and distinct things is the right and proper way of the world and better than a confused blurred mess of "intellectual property".

Truly innovative ideas: Patent and 20 years protection at some cost and significant paperwork.

Duplication or derivative of an expression (whether source code or binary files): Copyright and a long protection (too long but that is a separate topic) of that expression and derivatives of it.

Creative and unique design: Registered design (Design patent in US) fairly weak and narrow protection, not sure of the duration.

Copyright should be narrow protection as this case supports. That doesn't mean that no other protections should be applied to creative works.


In Europe one cannot patent shopping cart.


In Soviet Europe, shopping cart patents you!


Finally, someone did this! :D

(easy on the downvotes, it's just a little bit of humour, people)


It's not new that black box reverse engineering is allowed, I was taught this in my IT classes (I'm in the UK).


Indeed however in the UK we operate a common law legal system which means that precedent and case law account for much of our legal system (in effect the law says X is legal but then the case law defines exactly when and when X is legal).

A case like this is highly useful at staving off spurious law suits and spoiling tactics (particularly from bigger players against smaller players).

I think it's a good decision.


Note that this only sets precedent for cases under English law (and not Scots or Northern Irish law).


True, but it's worth remembering that the EWCA appeal was at least partly applying a ruling by the CJEU about the same dispute (C-406/10 SAS v WPL), which applies EU-wide.


The usual references for that are from the US though (the main one being Compaq's emulation of IBM's BIOS to make the first truly "IBM Compatible" PCs) or cases that were settled out of court so therefore have little impact on future cases.

This gives the practise a fixed definition in UK case law rather than people just assuming it is OK.

Of course this is only covering copyright: patents will complicate the issue.


> "In order to try to limit who can access learning or development editions of software products, companies may want to think about restricting who is the 'lawful user' of their software,"

And that's why we have 416 page EULAs.


The idea described in that part of the article, that software vendors selling their software to an organisation should try to licence to an individual from that organisation personally, also seems like an unhealthy principle to me.

If I'm an employee, I don't want to be tied down to some third party's arbitrary commercial licensing terms just to do my job for the employer who is paying for that software so I can. Software vendors put some crazy things in EULAs, and there's no way I want to be the guy going to court to find out whether any of them stands up when tested.

As an employer, it's an even more blatant cash grab than moving one-time purchases to subscription arrangements without any useful improvement in other areas to justify it. It creates a fixed cost that walks out the door with the same notice that the employee in question gives to leave their employment. And depending on the software vendor's policies, it may be an irreplaceable asset, if for example they force you to upgrade to the latest version when buying a replacement, which in turn creates built-in obsolescence.

I'm pretty sure I'd refuse to accept the terms of such a licence agreement in either role. There is absolutely nothing in it for the licensor, as far as I can see. It seems like it's just warping the concept of copyright another step in favour of the copyright holder, which the very idea of licence agreements already does to a dubious extent anyway.


I'm quite certain that most jurisdictions in the United States have some sort of laws to prevent the restriction of "clean-room" reverse-engineering... which is why projects like GNU and BSD were even possible to begin with. In any case, this is a huge blow for the big companies and a huge win for the smaller ones.


BSD was not "clean-room", and AT&T sued when it was freely distributed, because they didn't believe that all AT&T code had been removed.

https://en.wikipedia.org/wiki/USL_v._BSDi


Rivals may create copycat software through testing and user manuals?


I'm slightly surprised to see the statement:

"...non coding structural elements of software are not protected by copyright"

as I had understood that there is database copyright. Reference on the same site: http://www.out-law.com/page-5698


This decision was interpreting the Software Directive and the Information Society Directive. Databases are protected under their own directive (http://en.wikipedia.org/wiki/Database_Directive).


What about for the visual aspects of software?


In the U.S., visual aspects of software are generally copyrightable. Check out the Pac-Man v. KC Munkin case from 1982. There are several others. www.copyrightcodex.com/infringement/16-infringement-substantial-similarity/software-copyright-infringement#Pac-Man_v_KC_Munchkin

But when "visual" includes GUI, there's less copyright protection (because user interface elements are functional). Check out the Apple v. Microsoft case from 1994 as an example.

I'm not sure what the equivalent UK law would be.


Those cases are US cases. Given this thread concerns UK law, it might be better to find equivalent UK cases that establish this. (Indeed, the lack of previous UK cases is why the threads link is so notable)


You're totally right (I edited original comment to reflect that those are US cases). I'm not sure what the equivalent UK law would be.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: