Re: [Sheflug] Draft Computer Implemented Inventions Directive Me

[Bob Holland <bobholland [at] xxxxxxxxxxxx>]

On Saturday 11 December 2004 20:19, you wrote:
> Patents were developed so that methods and knowledge about how things
> worked could be publicly disseminated without affecting the inventors
> ability to make money from the invention.

Indeed - 'We will give you a monopoly if you publish how you did it'. The 
statement 'It's patented, so I can't tell you how it works' is an oxymoron 
which normally gets a resounding 'b******s' response from me.

Sorry for the diatribe below, but I am trying to get my own thoughts 
clear on this issue ready for when I go down to London tomorrow - comments 
would be most welcome.

In particular if you can point out any holes in my thinking then please do. I 
would not wish to waste time citing examples if there are gaping holes in the 
arguments that a Clapham Omnibus can be driven through - I am neither a 
politician nor a patent lawyer.

I have been doing a bit of research over the last couple of days. The UK 
patent office claims that it has not and will not (under proposed changes) 
grant patents for pure software but only for technical inventions which might 
be implemented by computer software - a position it claims it has adopted for 
many years. I thought I'd check out a particular patent of interest to me 
that might contradict this as an example to put forward - please see what you 
think....

US Patent US4617643 (full text is available on-line) published in 1986 with a 
filing date four years earlier on October 29 1982 reads...

<SNIP>
Syntax error correction method and apparatus 
--------------------------------------------------- 
Abstract

A digital computer system in which a series of programming series (sic) are 
processed and in which the structure of the programming statements must 
conform of a predetermined set of syntax criteria to permit processing. A 
programming statement is compiled in response to a computer input and 
displayed. The structure of the statement is examined in accordance with a 
predetermined statement of syntax rules. On the display selected color 
indicia are assigned to each character within the statement when it is in 
accordance with a predetermined set of syntax rules. The computer processor 
determines whether any characters remain to which color indicia have not been 
assigned and designates the statement as either acceptable or unacceptable by 
providing different color indicia bearing characters. In addition, the system 
includes a method of overlooking syntax errors to reduce operator frustration 
during running. 

Claims

1. For use in a digital computer in which a series of programming statements 
are processed, and in which the structure of programming statements must 
conform to a predetermined set of syntax criteria to permit processing, a 
method of syntax examination and correction comprising the steps of: 

compiling a programming statement in response to user input; 

displaying said programming statement; 

examining the structure of said statement in accordance with a predetermined 
set of syntax rules; 

displaying selected color indicia of each character within said statement 
which is in accordance with said predetermined set of syntax rules; 

determining whether any characters remain to which color indicia have not 
been assigned; 

examining the color indicia bearing characters alone in accordance with said 
set of syntax rules and designating said statement as acceptable or 
unacceptable, and, if designated acceptable, entering an altered programming 
statement thereof into the program formed solely of color indicia bearing 
characters. 

...more claims follow.

</SNIP>
------------------------------------------------------

The applicant was toy company Mattel Inc(US). 

Now as I read this, it is about using coloured syntax highlighting in the 
context of programming a computer - a technique now very familiar to us all.

Is this a broad 'software patent' or is it a 'computer-implemented invention' 
in the sense of the proposed legislation? If you believe it is the former, 
read on, otherwise bounce this back to me with your reasons why you believe 
it fits within the proposed legislation and I will shut up as I have clearly 
missed the point.

There are several issues here.

1) It has been awarded a patent in the UK - GB2131986 - and several other EU 
states DE3339270, ES8503869, FR2535489, NL8303721 and many others worldwide. 
No doubt Mattel thought it significant enough to spend a lot of money on it - 
maybe over $100,000 + annual renewals.

If you believe this to be a broad software patent of the form that the UK 
patent office says it does not and has no intention of granting in the future 
- then here is a clear example of the patent office doing exactly the 
opposite.

2) There was a period of 4 years between the priority date of Oct 29 1982 
(from which protection is applied) and the publishing date of 14 Oct 1986. 

Why so long? This is one RMS's arguments. Over this period of four years, 
anyone incorporating syntax highlighting would have been unwittingly guilty 
of infringement.

3) The author's of the Mattel patent could not have known other than by an 
exhaustive search outside their field of work that I had incorporated these 
techniques at least a year earlier in the control system of an automatic 
dough mixer. It formed part of what would be referred to now as an integrated 
development environment. Different parts of the language (operators, 
operands, input/output functions etc) were rendered in different colours. 
Anyone of the 10 or so engineers who were developing the application software 
on that project at the time will remember this as a useful aid to programming 
the system. The machine was first on the UK market in 1981 and launched in 
Las Vegas during 1982. Its introduction to the baking industry is well 
documented in a book called 'Breadmaking, The Modern Revolution' 2nd edition, 
again published in 1982. I even demonstrated how to program it to the South 
Yorkshire Personal Computer Group (gosh - I have been around that long!).

I am not claiming here that I 'invented' coloured syntax 
highlighting - far from it - merely that my use predates the Mattel patent. 
Surely this is a 'concept' not a 'technical invention' or am I wrong? I also 
note that I may well have unwittingly infringed existing patents by 
having an independent idea and implementing it - but surely this is a 
copyright not a patent issue and I definitely did not copy anything.

This illustrates another of RMS's points - that prior art searches 
are woefully inadequate when it comes to software and always will be since 
the possible fields of endeavour are so numerous.

Note that as far as I can tell, many major companies have licensed this 
patent which presumably means you, me and many others have been paying for 
this ever since. I only discovered the patent a couple of years ago when I 
did a search on the web after thinking how useful 'my' idea had become.

4) My implementation of the coloured syntax highlighting was very simple once I had had the idea - involving little more than 
the addition of teletext colour codes to the language parsing algorithms. I 
can certainly recall the steps involved in doing this - even down to where I 
was sitting at the time. It was little more than a few hours work between the 
idea and a working implementation. It was not regarded by me as a significant 
inventive step - just an obvious and useful extension of what I was working 
on at the time. Even though I was pleased with the result the idea of 
applying for a patent never entered my head and (most importantly) as a small 
company we would not have had the money to consider this anyway.

This goes back to (2) above. If what might appear to 'the patent system' to 
be a significant 'invention' can be implemented from an idea in a few hours 
compared with the patenting process taking four years, is the patent system 
itself the most appropriate IP protection vehicle for this type of work?

It seems to me that the system is fundamentally flawed in this respect.

5) Another case...

A few years ago the UK patent agent working for my employer at the time 
suggested patenting a computer-implemented 'graphical display of a sensor 
reading as a plot against time'. The line was 'you inventors/engineers always 
fail to recognise when you have invented something'. I believed the so called 
'inventive step' was all about 'not being obvious to someone skilled in the 
art' - but apparently the real interpretation is more about 'not being 
obvious to someone (a)skilled in the art and (b)incapable of original 
thought'.

The GB patent was granted - much to my surprise. Admittedly this did have 
some industrial benefit to our field of work but nontheless the 
principal claim was the trend display. Cost to the company for the UK-only 
patent (a stupid thing to do anyway) was ~£7000. Who really benefitted from 
this - the company or the patent agents?

This implies that anyone displaying a graph of a sensor reading against time 
on a computer might technically be guilty of infringement. What about using 
an oscilloscope to monitor the same signal then? This seems absolute rubbish 
to me and I still can't understand how it was possibly granted.

Would any of you even consider that you might be caught out by this or the 
many similar patents which might exist in the UK during your work?

Would this patent - already granted in the UK - still be allowed under 
proposed legislation? I guess so - since the UKPO is saying it is not 
changing its position and has not allowed broad software patents to be 
granted in the past.

I am also named on two similar patents related to signal processing - so 
watch out what you do (especially with concrete) or someone might come after 
you!

To summarise then...

I believe there is evidence that the UK patent office and other EU states 
do in fact award patents for software applications which are broad in scope. 
The syntax highlighting example is surely by definition a software patent. 
This is contrary to the stated position of the UKPO.

The patent system is totally incapable of thorough examination of prior art. 
I believe my work on syntax highlighting would have been highly relevant to 
the Mattel patent, but it is most unlikely that anyone involved in that chain 
of events would have uncovered it during examination.

The timescales involved in the process of patent examination are incompatible 
with the pace of technological development in many fields, therefore calling 
into question the benefits of the patent system.

In a worldwide market, the costs of obtaining suitable patent protection rule 
out SMEs and individuals. Even just covering the main technological countries 
will typically incur costs of £25,000 upwards for initial filing, let alone 
renewals. The best course open to SMEs might therefore be to (a)either keep 
their work secret, which many do, knowing that a competitor might obtain a 
patent at a later date which might then involve a trip to the courts or 
cowering submission; (b) to publish their work in some obscure journal to 
prevent a patent being awarded elsewhere; (c) seek venture capital to afford 
the patent (but there is a chicken & egg situation here). In none of these 
cases does the patent system benefit the SME.

If drawing a graph of a sensor reading against time on a computer screen 
infringes a patent, it will become impossible for anyone to work without 
looking over their shoulders.


Regards to all,
Bob Holland





































 
___________________________________________________________________

Sheffield Linux User's Group -
http://www.sheflug.co.uk/mailfaq.html

  GNU the choice of a complete generation.