Software developers are angry that Microsoft has been granted a patent for the conversion of objects into XML files.
The patent, which was granted by the US patent office on Tuesday, is for XML serialisation and deserialisation — the conversion of a programming object into an XML file and vice versa.
XML, an open standard developed by the W3C, is commonly used for representing data structures in applications that exchange data. As applications will often need to convert between programming objects and XML files, this patent could cover virtually any application that uses XML to transfer data.
Wookey, a developer on the free Linux distribution Debian, said the patent should not have been granted and could be used by Microsoft to discourage competition.
"It's an abomination to anyone who does programming that the concept of every possible way of converting between a programming object and an XML file (and vice versa) can be owned by one company," said Wookey to ZDNet UK. "Should someone else want to save a programming object in the form of an XML file, Microsoft can now charge them for the privilege or simply refuse them permission to do it at all in the US."
Michael Tortolano, a senior software manager at Home Media Networks, agreed that the patent should not have been granted.
"Interoperability is the life blood of the computer industry yet we keep seeing software patents being granted which directly threaten the ability of computers to their basic job, to share and manipulate data," said Tortolano.
Microsoft had not responded to a request for comment at the time of writing.
Although such a patent is less likely to be granted in Europe, which has a less liberal patent system than the US, there are concerns that this could change if the European software patent directive is adopted by the European Parliament. Opponents of this directive, which is due for a final vote in the summer, claim that it will lead to the widespread patenting of software in Europe.
Jonas Maebe, a spokesman for the Foundation for a Free Information Infrastructure, said that data conversion patents have already been granted by the European Patent Office (EPO), for example, SAP was granted such a patent last year. This patent, EP1267277, is described as a method for "modifying the data structure used by an application program to access database systems in a computer system".
"This [Microsoft patent] once more shows that the practice of the US and European Patent Offices is quite similar," said Maebe. "It is clear that by codifying this EPO practice in a directive as the [European] Commission and Council want to do, they will only make this sort of US-style patents enforceable in Europe as well."
The Microsoft patent that developers are concerned about is patent number 6,898,604 , which was filed in June 2001.
Talkback
Where's the liability for awarding wrong patents?
there is none
this spells the end for XML
XML was born as a particular format for data storage.
A programming object is made up of data.
Where is the patentable "innovation" in using XML for the purpose it's born for?
It just seems people who review patents at patent offices seem to have no skill sets at all when it comes to IT. When you see at the increcble amount of crap that gets approved...
Some people should fired!
Whatever happened to prior art?
I worked as a patent clerk when I was younger, and most of our Patent Agents spent extreme amounts of time searching for conflicting Patents, or ensuring that the thing they were attempting to Patent could not be claimed to be obvious and in general use.
Whats with the US patent office, do they just take the money and rubber stamp everything these days? (they never used to)
To be fair to the patent office in this case this is a patentable invention in the US. They are just applying the rules.
And it is very difficult for them to search the prior art when nearly all of that lies outside the patent database. It may well be that no one had got round to specifying a mechansim for this particular data-structure<->XML conversion when the patent was applied for, so maybe there wasn't any exact prior art.
My understanding of the system is that prior art for some other similar data-strucutre<->XML conversion generally doesn't count. Which is particularly stupid in the software area because any such conversion is much the same in principle as all the other possible conversions and it's incredibly obvious to programmers that if you wanted to serialise a programming data-structure you'd do it just the same way as you did for serialising spreadsheet data-structures, because software is intrinsically generic. That a big part of the reason why the patent system fits it very badly.
So the only thing this patent could technically be failed on is obviousness. But in fact patents are very rarely refused on obviousness grounds because there no such thing as 'too small' an inventive step. Effectively if it really is even a tiny bit inventive that's good enough for the patent office. It seems to be an intrinsic problem with the system as people have been trying to raise the inventive step for decades with negligible success.
This is why you should never believe patent attorneys who say "there is nothing wrong with software patents per se - just the ones that are too trivial or broad - we can fix that". They are lying or mistaken - they can't fix it, and that's a good reason to keep software out of the patent system in Europe. Even if if you don't object to them on principle, you can just object to them in practice.
The patent system in the US is destined to become a laughing stock with another 10-15 years of this sort of thing. It'd be nice if some sanity could be injected before _too_ much damage is done. It'd be even better if we could laugh at them here in Europe because we managed to get some relatively sensible law. That won't happen unless people - especially SMEs, talk to their MPs and MEPs about this _right now_. The current directive will give us _exactly_ the same environment as the US, which sadly is what the leading representatives of the Conservatives (Malcolm Harbour) and the Lib Dems (Sharon Bowles) are fighting for.
The link in the article points to the wrong patent. This link works for me:
http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PALL&p=1&u=/netahtml/srchnum.htm&r=1&f=G&l=50&s1=6,898,604.WKU.&OS=PN/6,898,604&RS=PN/6,898,604
If this link is not usable for you, this procedure should work:
browse to http://www.uspto.gov/
click on Search
click on Patent Number Search
Enter the patent number 6,898,604
Click on Search
Read
Throw up
The United States Patent office merely takes the position that it is a matter for the courts to decide. A patent is worthless if it can not be defended in the courts. Of couse that would cost some little guy a lot of money. But if that little guy has been supporting Microsoft all these years; to bad and too late.
The cure for spam is to make it uneconomical, like dont buy the stuff. The cure for microsoft is the same.
We should feel sorry for the easy way out slobs (you and me) that bought Microsoft when there were other choices.
A Micro Soft is what you get after an exhaustive session of mankinds favourute activity (well for males definitely). Microsoft is proof that your money in the right hands can fix anything and it is also proof that 95 % of the poulation are asleep, and the developer community is in REM sleep.
At least Gates gives some money away, he knows he can't take it with him. Which is more than can be said for the rest of the commercial/proprietary software industry. Crawl into a hollow log, Microsoft is on the prowl. There are more than two sides to any coin in cyberspace. (sorry about the typos.)
I am mousseau@magama.ca
I am sorry to disagree, as most people don't care. Remember, if it isn't illegal, then it is allowed. Your assumption that the small effect of your boycott would cause a rethinking of your policy is laughable.
Companies like Union Carbide (now Dow Chemicals) were and are unpunished (ok, a mild slap on the wrist) for the killing of tens of thousands of people in Bhopal India. The CEO was never extradited for trial, and a token sum was given ($470 Million with an inital death count of over 8000, which is now estimated to be over 30000 and still growing.)
So, you will stop buying microsoft products, but how many M$ products do you buy per year? I am sorry, but with billions in the bank and an addicted populace willing to pay like a crack addict they are never going to feel your protest.
Boycotts work when a community can be brought together. The bus boycott worked because a populace that were the primary users of the service joined together and stopped using. Microsoft will not illicite the same response, as the community is fragmented and the decision is rarly made on a personal level. I could have decided to not ride the bus, but a cube worker cannot chose not to use windows (The sale was made when the company bought the computer).
geez louise... they're expecting folks to just see this headline:
InfoPath 2003.A Global Solution for International Trade
http://office.microsoft.com/en-us/assistance/HA011878631033.aspx
- and ignore the "this" in "this implementation".. as in:
http://www.unece.org/etrades/unedocs/
"It is an open and technology-neutral solution that can be easily implemented by SMEs and large companies alike."
on the other hand:
ECM Report :: [OTEX] PureEdge and Open Text Partner on E-Forms Solutions
http://www.ecmreport.com/index.php/articles/news/171
microsoft people's participation in the working groups adds witness:
http://www.w3.org/TR/2004/REC-rdf-testcases-20040210/#ntriples
http://www.w3.org/TR/2004/REC-rdf-testcases-20040210/#ref_cwm
An N-Triples document is a sequence of
US-ASCII characters and is defined by the
ntripleDoc grammar term below. Parsing it
results in a sequence of RDF triples formed
from the subject, predicate and object
productions
http://www.w3.org/TR/1999/REC-rdf-syntax-19990222/#intro
"This specification defines two XML syntaxes
for encoding an RDF data model instance. The
serialization syntax expresses the full
capabilities of the data model in a very
regular fashion. The abbreviated syntax
includes additional constructs that provide
a more compact form to represent a subset of
the data model. RDF interpreters are
expected to implement both the full
serialization syntax and the abbreviated
syntax."
http://www.w3.org/TR/2004/REC-rdf-testcases-20040210/#ref_cwm
http://www.w3.org/TR/1999/REC-rdf-syntax-19990222/#acknowledgements
(microsoft participated in working *groups*)
MS is probably just twigging b/c the department of defense went with OpenText
Well, it's not the end of XML for storing objects just yet. Don't forget that this only applies in the USA (and Australia, which is bound by treatises to uphold US patents). Besides, having been granted the patent is not the end of things. All we need is someone with the resources to mount a patent challenge (for example, look at what happened to MS's patent on the FAT filesystem). Unfortunately this means that someone has to 1) actually obtain a copy of and read the patent, 2) Discover prior art which was public knowledge (woohoo - aren't you glad we have SourceForge?) and 3) write a letter to the patent office informing them of their mistake and asking them to review the patent with the evidence you present. Now if you're lucky the patent office will investigate and then invalidate the patent. If you're not so lucky the patent office will investigate and then call in Microsoft and your party in an attempt to sort things out - this part could be incredibly costly. If you didn't do a decent research job then the patent office can uphold the claim and you just made it tougher for the next person in line to challenege the patent.
isn't a software company supposed to make money by selling software rather tahn selling the right to make software?
Look at http://sourceforge.net/projects/sxp - isn't it a prior-art? First version was available in February 2000. I wrote a similar software but it was after July 2001 ;-)
Thanks for pointing that out Bill. I have updated the link in the article.
I worked for Bowstreet. We were serializing communications and programming objects into and out of XML before 2000. I don't know the particulars of the Microsoft patent but I'm sure I could scare up some design notes that might establish prior art.
Would this mean that C++ or other programming languages are being patent? Of course not! Microsoft SHOULD NOT BE AWARDED WITH THE PATENT!!
At this rate, I soon will not be able to initialize a variable without violating a patent. Or has that already been patented?
I believe I will patent the equally inventive concept of securing shoes to one's feet with shoestrings. I'll then license my technology to Bill Gates for 300 billion dollars.
It's a variable. Could represent anything. Depends how you initialise it, what you initialise it with. Might very well be infringing a patent. Best to leave it uninitialised :-)
I know the US patent system is based on grant first, argue and make lawyers rich second - but the spirit of the all patent law states that prior art must be taken into account.
Well Microsofts pathetic attempt to again patent the existing has violated that premise. There are numerous examples of prior art for serializing objects to XML from Sun's JavaBean XML serialization to SXP hosted on SourceForge. There are so many example including one I wrote myself about 5 years ago.
Unless patents can be enforced as they were intended, there is little point in them being used to prop up already mighty and agressive organizations - a complete violation of the intention of patents (to allow inventors to do something with their invention before the big boys try to stomp all over them like the school bulley).