Slashdot Mirror
Multilingual DNS Patent Roadblock For IETF
Xanni writes "Intellectual property claims have blindsided the Internet Engineering
Task Force and could derail the group's efforts to develop a common
scheme for supporting foreign-language domain names across the Internet.
NWFusion is carrying the story."
The NT line of OSes has always been 100% Unicode.
A file extension isn't needed: notepad utilises a BOM (byte order mark) to determine the file type. If you've done any work with XML, you will know that BOMs are used there too (although not mandatory). The BOM allows you to determine whether the file is UCS-2 or UCS-4, and also specifies the endian-ness.
I don't think MSFT has left anybody in the dark.
Developing NT only international products is actually very nice, especially when used on a system with NTFS. International stuff can interoperate so much more easily in this environment. Unfortunately, due to the limitations of Win9x and the need to support it, most builds are multi-byte only, not Unicode.
FYI: The Win32 API under NT includes ANSI versions of virtually every system call, giving compatibility with Win9x binaries. These ANSI versions do multi-byte to wide translations, and then delegate to the Unicode version (yes: big performance hit.) What is bad is that under Win9x, the wide versions of the functions exist, but they are normally just empty stubs that do nothing! e.g. if you look at the exports from user32.dll, you will see that MessageBox consists of MessageBoxA and MessageBoxW.
When these people, who sell Suzuki cars in Iceland, ventured on the web, they naturally wanted to use the name of the company, "Suzuku bílar" as a domain name (without the space, of course): "suzukibílar.is". But they couldn't do it. DNS doesn't allow it. So they did what is usually done, and replaced the acute i with a regular i. This is kind of unfortunate because "bilar" means "breaks down" or "malfunctions".
But I guess you don't see that as a problem. I mean, why can't these people just standardise on English?
--
Fuck the system? Nah, you might catch something.
Which countries are the ones that want internationalized domain names? Probably those which don't use ASCII. Which country was the patent granted in?
Couldn't the DNS servers be run in the countries where they are needed, where they won't be affected by the screwed-up US patent system?
-- Ed Avis ed@membled.com
But RFC2026 section 10.3.1 makes it pretty clear that any conforming submissions must disclose "... the existence of any proprietary or intellectual property rights in the contribution ..."
Is this merely intellectually dishonest, or is it fraud?
I can code, and think you're comment is ridiculous. This is a pretty obvious idea. If I were thinking of a way to encode foreign language names so DNS could deal with them, the solution they patented is probably the one I'd come up with too. The only work they've really done is run to the patent office and pay them a money.
Need a Python, C++, Unix, Linux develop
$10,000 is nowhere near the actual price of patenting something like this.
I just went through the process of pricing patents. A patent on a mechanical device was estimated in the tens of thousands of dollars - the actual patent fees charged by the USPTO plus the patent attorney's fees. An international patent was in the realm of six figures.
Challenging that patent will cost millions of dollars. Have you written your check to the IETF to help defray their legal expenses, or are you more of a "Let's you and him fight" kind of guy? Me, I'm more of a "What a bunch of IDIOTS!" kind of guy. Welcome to the desert of the real...
Isn't attending standard group meetings and patenting the ideas presented at those meetings a patented business model owned by RAMBUS?
I'm trying to teach myself to set people on fire with my mind... Is it hot in here?
Yes, but the patent was granted earlier this year, which means it was patented quite some time before.
If they hope to challenge it, they'll have to go with obviousness, which the patent obviously is. We did exactly what they patented to slap our i18n info into our databases, which didn't accept double byte characters. It's a trick i18n firms have been using for years, and, as usual, it's astonishing that someone could get a patent on anything this obvious.
Here's the heart of their patent: "The domain name is converted to a standard format which can represent all language character sets, such as UNICODE. The UNICODE string is then transformed to be in RFC1035 compliant format." Or, according to the article, "This solution involves converting foreign language characters into Unicode, a computer industry standard, and then encoding them in U.S. ASCII for transmission over the Internet."
There's an obscene amount of prior art for this, applied to databases, dll's, and pretty much anything you can think of, except DNS. Even the courts usually don't allow a patent to stand that's simply old hat applied to a slightly tweaked problem.
Prior to this patent coming to light it had been suggested that it was time to do something different with regard to DNS.
John Klensin, the chairman of the IAB, presented to the IDNwg in December, at which time he dropped what I would describe as an intellectual bomb, by saying that the international train had already left the station, and that IDN shouldn't hurry to get a temporary solution or incremental fix out the door. Instead, he said that the IETF should strongly consider a real directory-based solution. He's probably right.
While it is patently obvious how to encode domain names, what someone in a foreign country would see is gibberish.tld. That's not what you want someone to see. You would rather have people see your domain name translated into languages they understand.
Of course, a real directory would be nice, but we've tried it several times before, and without much success. Remember about two years ago, directories were all the buzz? Now where are they?
Also, as most here know, this is an area where politics and economics really make things ugly. So no matter what system we come up with, we'll see companies like Verisign do their best to winning another land grab.
This is starting to resemble the initial dot-com fever as swept corporate America. Anything that had those magical 4 characters in the name was new in an exciting way and so must have been a sure shot. We know what happened there.
The new craze is this IP (two NEW magic characters!) suit stupidity. Any possible patent is sought out and somehow granted, just to say "We were first, so there." It's not even so much about licensing fees as revenews, which most of these attempts have never generated in any significant form. This is about having control, being able to swing a legal club and act big.
Won't it be nice when corporations have a clue again? I'm sure by then there will be another way of feeling all powerful and amazing after this one wears off so I guess that's still wishful thinking. Technology makes things different and on many fronts easier - but no portion of it is miraculous. You still have to work
In other news today, stupidity is rampant on the internet.
Any spoon would be too big.
The big problem with your reasoning is that it it's a double-edged sword. The difficulties you're predicting as an English speaker are exactly the difficulties that non-English speakers are trying to deal with right at this moment.
For instance, I'm of Japanese descent. I can only imagine the difficulties of some older Japanese people who know no English, when they have to type in domain names.
"Microsoft," for example, would be written phonetically, but it would come out like this in Japanese:
Even after having described the English alphabet, how do you explain that the mangled name above (it's the closest you can get in Japanese) is to be written like this:
Want an even worse example? Try slashdot:
Actually, that's too phonetic. If I wrote the equivalents of the actual Japanese letters, it would come out:
Don't ask. It's a quirk of Japanese spelling.
--
Accountability on the heads of the powerful.
Power in the hands of the accountable.
They patented a method to convert text from an arbitrary charset, to UNICODE, and then convert this to plain text as U+XXXX where XXXX is the hev version of the unicode charset. Thats a basic first year programming assignment and definately not an innovation. The first part is obviously crap as converting from one charset to another is not a new idea (Hell java does it for nearly every charset in existance for years). As for the rest does outputing a unicode character in hex count as an innovation? Is there anyone here who works with unicode and hasn't done the same thing as part of debugging an application?
Whats next, patent a method for converting from an arbitrary number set to decimal and then outputting it in hex? The stupidity of the USPTO boggles the mind.
Slashdot: Proof that a million monkeys at a million typewriters can create a masterpiece
As usually, the patent is also available at US Patent office
I'm still trying to figure out what people mean by 'social skills' here.
Good idea! I'm glad I thought of it. Now, if you'll excuse me, I need to make a quick trip to the Patent Office...
(and while I'm there, maybe I'll trademark "DNS 2.0")If the IETF has been working on this solution for so many months, they must have some evidence to show for it. Couldn't they just cite their own research as prior art and have Walid's patent tossed out?
And on another note, who exactly would pay the licensing fees that Walid is asking for?
Dancin Santa
This is a perfect example of why you should not be allowed to patent software. Here's a group trying to create a more universal internet, and they get knocked back by a software patent. This is a prime example of patenting software and stifling innovation. I wonder how much sooner the internationalization would be complete if not for this roadblock.
Where's my lobbyist? Right here.