Slashdot Mirror
Microsoft Seeks Latitude/Longitude Patent
theodp writes "Q. What does Microsoft feel is unpatentable? A. Apparently nothing! On Thursday, the USPTO published Microsoft's patent application for the Compact text encoding of latitude/longitude coordinates, in which the software giant explains how a floating-point number can also be represented as a less-precise integer that's displayed in base-30 notation!" If ever I have seen a silly patent, this is it.
Pretty sure the US Patent Office has a say in what is and isn't patentable.
Support the First Amendment. Read at -1
Who care's what they patent? There's no chance in hell this would stand up in court, the judge would laugh it off and Microsoft's patent gets revoked.
Is there prior art? Probably.
Is it obvious?
Probably.
Is it capable of winning a course case? AAHAHAHAHAHAAHAA!!!
End of story.
Don't you think the title is a little trollish? I get it, already. MS is evil. But they're not going to deploy interceptors to stop ships using LAT/LON coordinates while out at sea.
Wait... maybe they will...
If Nalgene water bottles are outlawed, only outlaws will have Nalgene water bottles.
Tomorrow, they'll add an amendment to that patent saying that they own the location of that coordinate too, and if granted, Microsoft will own any piece of land they want. MS: We own this coordinate. Linus: But this is my house. MS: Not anymore. Linus: I'm homeless :-(.
This is a problem with USPTO, not Microsoft. I mean, how is this any different from me downloading pirated movies? It's wrong, but I do it anyway because I can (get away with it). It's the same with Microsoft, they can patent pretty much everything because the problem lies with "the system". Fix the USPTO, fix the problem. Fix the distribution system for movies, fix the warez problem.
I'm usually pretty lenient about what I consider a good patent, but this one seems dumb. It's not even an optimal way to do it. They're just using a fairly cheesy form of compression to make a string shorter.
So, they remove the ability for a human to tell what the lat/long is by inspecting the string, but compress the string suboptimally? Whiskey Tango Foxtrot?
Pulp Audio Weekly - Geek News and Reviews
What happened to not being able to patent a mathematical formula? Isn't something like this basically just a math trick?
Manufacture in China
This only confirms what we have all known for a long time, Microsoft is run by a bunch of morons.
For a "bunch of morons" they seem to have done a pretty good job establishing and maintaining desktop and office suite dominance.
For a "bunch of morons" they seem to have made a pretty big warchest of cash.
For a "bunch of morons" they seem to have gone from nothing to second place in the game console market rather quickly.
For a "bunch of morons," in other words, they're pretty damned successful. The last thing you or anyone should be doing is writing them off.
The coolest voice ever.
I know it's in your blood to hate Microsoft, but least take the time to read the patent! What they are seeking to patent is a METHOD of encoding LAT/LONG in a URL in a better way than is currently employed. I think this patent is incredibly valid.
A blog like any other.
From the patent abstract:
"Methods are disclosed for encoding latitude/longitude coordinates within a URL in a relatively compact form. The method includes converting latitude and longitude coordinates from floating-point numbers to non-negative integers."
Where are tinyurl and similar websites to claim that they have been converting URLs to relatively-compact-form, using non-negative integers and letters?
Although M$ churns out mundane but fairly good software as the main line of business, this company has a huge R&D budget, and the central research laboratory at M$ is the equivalent of the old Bell Laboratory. I suspect that the M$ laboratory is not confined to only software research. There is likely small, upstart efforts exploring other technologies.
Certainly, X-Box took me by surprise as, up to its debut, I had always thought of M$ as a software company.
May be, there is something to the warning: "Resistance [in any technology] is future. You will be assimilated."
"The set deliberately omits vowels to avoid the possibility of the algorithm inadvertently generating real words that could be offensive."
Wow, that is SO "politically correct". Still it does't prevent people from constructing URLs saying fvck 0ff. It would be better if people would simply learn to respect other peoples freedom of speech.
Visit http://ringbreak.dnd.utwente.nl/~mrjb/growingbettersoftware to download your free copy of the book
At least one part of this encoding method - converting floating-point to non-negative integers - looks like prior art.
For instance, ESRI does this for their spatial database (SDE):
URL is here.
And most likely it was known even before...
This type of patent is NOT about protecting their rights to an innovation. It's about restricting interoperability.
After patenting this encoding method, they can create some kind of software interface based on it, e.g. have a web application that uses this encoding in it's URLs, or an extension to Internet Explorer that uses the encoding somehow. Then if the server/extension becomes popular, they can use the patent to lock out OSS and other vendor's applications.
Location-awareness is a hot topic these days -- that probably has something to do with this particular patent.
Of course, if there were no software patents, then Microsoft wouldn't need to do this.
I suspect the real reason for this is so they can control/prevent deep linking into their Terraserver (etc) geographical systems. If my website has a way of generating their coordinate URLs and linking directly to their content bypassing their front page, they could now prevent me from doing this because of this patent.
Jolyon
Please read my Canon EOS tech blog at http://www.everyothershot.com
Don't you find it more believable that they are simply waiting for software patents to be established world wide, after which they'll take out all the major open source applications? I.e. Samba, Apache, Open Office, Mozilla, maybe gcc, etc.
The point of encoding lat/lons this way is to allow a compact coding in URLs.
That's useful in mobile devices, where URLs are limited. It's also useful in that now you might be able to memorize and type in your latitude/longitude, since in a higher base (30 is just an example) you can get good precision in few digits (their example is 5 characters, with 2-meter precision).
Why base 30? That's 10 digits and 26 letters, minus 6 vowels "to avoid the possibility of the algorithm inadvertently generating real words that could be offensive". Funny.
So it's useful. As far as I'm aware nobody's ever done it before, which makes it both non-obvious and novel. Those are the three tests of a patent. If you don't want to use it, keep using base 10. If you do want to use it, at least give Microsoft credit for coming up with a reasonably clever idea. As another poster pointed out, this is the type of patent MS usually uses defensively, so that nobody goes out and patents an idea they're already using in live software.
I think MS would like to see everybody memorize the lat-lon of their home as two five-digit strings, but it's not going to happen. First of all, the patent requires you to pick a precision beforehand; decimal degrees and degrees/minutes/seconds don't require that. Second, even if MS introduced a standard, they'd better release the patent for public use, or nobody will bother lest they risk being sued. Decimals are wordy, but everybody understands them and they're free.
I have one other gripe about the patent. They spend considerable time explaining how to convert a number in base 10 to a number in base N. It's not one of the claims, and it really could have been taken as given.
The headline is misleading, becuase they are not patenting lat/lon, just one method of representing it in a URL.
I do think there are a number of holes in it. For example, claiming a patent on the method of concatenating 2 strings together just because they generated those strings creatively to represent coordinates.
Wrapping up, I think the patent is valid, but is a mix of patentable and non-patentable statements. It is at least an interesting algorithm to study.
Why, oh why, didn't I take the Blue Pill?
(sheesh, the arrogance displayed on Slashdot knows no bounds)
Sheesh, the lack of humor displayed on Slashdot knows no bounds.
Infuriate left and right
Second, I think there's prior art possibly in the universe simulator Celestia which supports URL encoding of coordinates (I don't know if it's uses the lat/long system though, that's why I'm bit unsure), and there definitely seem to be prior art in NASA's World Wind application. It uses a compact Lat/Long => URL encoding scheme as follows:
Beware: In C++, your friends can see your privates!
Patent application in its wording uses conversion to compact ASCII string.
So, you can avoid infringement of this patent by using conversion to ISO-8859-1. Or perhaps a new patent, anyone? Just leave ISO-8859-2 for me, please.
There you are, staring at me again.
Just because no-one's ever done it before doesn't mean it isn't obvious - it usually means no-one's felt the need before. If you came to me and said "Paul, we need a compact way to represent latitude and longitude in URLs, what do you suggest?" this is exactly what I would have reeled off at my desk without even needing to pause for thought, modulo the rather silly thing about leaving out vowels. I'm sure the same is true of every programmer in my workplace.
MS aren't introducing a standard, quite the reverse - they are trying to prevent people interoperating with their servers.
Xenu loves you!
0392670 and 1416000, seven digits each. You concatenate them together in a base-N alphabet. So if in base ten you have 03926701416000, nothing gained except I would like to know what is at that digit of pi maybe but no real use regarding the patent.
You could use a websafe alphabet (like I use when encrypting form data between one page and the next, based on a public CPAN module.. encryptform or some such) or a little bigger alphabet that would be MIME or Base64.
From item 8 they are dropping accuracy in order to encode in less characters. Um. Well yes you can shorten numbers to lose accuracy. If you write the numbers using letters instead, like in base 16 or some substitution alphabet it may look like you are shortening a word but really it is just dropping decimal places. Microsoft claims they are unique at being able to go back and forth between string length and allowable error. They have a patented subroutine that you feed say a floating point latitude, number of characters to use, the number of characters in the alphabet (i.e. the base) and required accuracy, and it will spit back something like "KXW" maybe.
Likewise you can feed another patented subroutine "KXWCMY" and it will give you back something like "39.3N, 142E" (well higher res than that really, it doesn't seem that useful unless you are measuring GPS coords to the inch). Perhaps this is the code a mobile device will shout whenever it can triangulate its location from a few known wifi points. :)
Well I just skimmed the end of it but it seems this is for use when you really don't want to use all those decimal places (8 digits for meter resolution). Needless to say 32 bits is enough to handle it but it looks so *long*! So instead of just lopping off the last few digits, they want to compress it (okay so far) but then they tell the compression algorithm how compressed they want the string to be, how much they are willing to give up (I would think in decimal places but ultimately in meters I suppose).
They then talk about personal info managers and map display programs on pdas, and the bs starts to pile up real fast. They start talking about nonvolatile memory, video tape, scanners, joysticks, office environments, what have you.
There is mention of an URL (301) that 'contains a geographic parameter "mapcoord", which has a parameter value "ry7cx4tp95"'. There is some talk about users inputting information which sounds interesting, until you realize that in the end this is really a quintessential rot13 for the 21st century, written by a corporation that does not care if users cannot decipher the codes or tell how accurate it is at a glance, or find it on a globe or non-M$ map, that assumes every gps manufacturer will liscense the patent, who cares if you don't have alpha input on your keypad etc. Someone should tell them you could do it all in just a couple characters on a kanji-equipped Japanese phone. While it gets more seductive as you read more and more, it also hits you with a sledgehammer that you have to have a calculator with the patented subroutines built into it, just to understand what codes your are typing.. it can only ever be useful among a weenies who have been brainwashed to think in corporate speak and that is the problem with Microsoft and Windows. If they just published it for free openly most people would forget it (it seems neat maybe but in the end it's just too much trouble unless it is an accepted standard like geo8 for an 8 letter string.. and even then). As it is I think it is utterly disgusting. Also it is probably beaten by error checking code, lossy image compression code, and the CPAN module I mentioned. Yuck!
Look at: http://www.tapr.org/tapr/html/Fmic-e.html
This looks quite similar: http://www.nacgeo.com/nacsite/documents/nac.asp/ ... right down to the base 30 representation.
Try it out:
http://www.travelgis.com/geocode/
This co-ordinate encoding scheme sounds similar to one used by US (and other) military forces for air targeting.
See this description of GEOREF co-ordinates for example. Basically you divvy the world up in to a grid and use letters to reference the major fractions of the co-ordinates and numbers the minor fractions. So 106 25' 44" W 310 48' 06" N becomes EJPB 3448.
.sig?
Come on, it's legitimate, it is innovation that supports their trademarked slogan. ;)
The string encodes Lat, Lon, nature of the event and control digits to minimize errors.
Some characters are excluded, but not exactly vowels. The idea here is to prevent misreading (eg "O" (letter) and "0" (number). Or "l" (letter) and "1" (number) and so on). So the base used is not 36, but something around 30 (can't remember precisely).
Sorry, once more with paragraphs
So there have been 150,000-300,000 software patents granted since 1998? Or is that merely when the appeals process for the lower court ruling ended?
The problem I have with software patents, David, is that such a very small percentage of software is written for sale. Most is written for use. The USPTO seems to have no reasonable way of knowing what techniques have and have not been used by some programmer somewhere in the country. They can look at previous patents (there were none, initially), they can look at journals, but there is no viable way to prove lack of previous invention and use in someone's in-house code development. There are just too many programmers doing too much work.
With 150-300k software patents out there, with some high fraction non-novel or obvious to experts in the field, programming has become a minefield.. even things that were nominally novel due to context, like writing software to do an old function on the Internet, are being locked away, and by companies that themselves benefited tremendously due to the use of software techniques developed and shared freely before software patentability.
How does anyone but Microsoft benefit by Microsoft patenting a wide swath of approaches to reducing Spam? Microsoft gained many tens of billions of dollars of additional revenue due to the surge in PC sales brought about by the widespread adoption of the Internet, yet they and others are now putting roadblocks up as fast as they can to control any further elaboration of the Internet, or competition in providing compatible services on it.
How can this possibly be defended?
- jon
Ganymede, a GPL'ed metadirectory for UNIX
Okay, let me ask you some questions:
Let us take, for example, the latitude 36 degrees, 1 minute, 30 seconds. (Assume WGS-84 for the entire post.)
If I convert it to 36.025 degrees, is that a new projection?
If I convert it to 3.6025*10^1, is that a new projection?
If I say it in Spanish, or any other non-English language, is it a new projection?
If I write it thirty-six degrees, one minute, thirty seconds, have I made a new projection?
Now let's define a new unit, call it millidegrees. That is, one thousand millidegree = one degree.
If I write it as 36025 millidegrees, have I made a new projection?
Now let's take that number and write it in hexadecimal, 0x8CB9, is that a new projection?
None of these have made a new projection (or projection onto a plane at all), because we are taking the exact same geographic coordinate (forget about not having expressed a longitude for now, the same argument applies) in DEGREES LATITUDE.
In fact, we are still talking about spherical coordinates, are we not?
Now, let's take one of the above numbers, 36025 millidegrees, and convert that to Microsoft's base 30, and we get 1B0T millidegrees in Microsoft base 30 (the set of digits does not use vowels).
1B0T (base 30) millidegress = 36025 (base 10) millidegrees = 36.025 degrees (which is what Microsoft says the original coordinate is expressed in).
If we take both a latitude and longitude, and convert them into Microsoft's representation, then we have not made a new projection because:
(1) We have not changed the model used.
(2) We have not performed a transformation on the coordinates other than changing base. Particularly, we are still dealing with (spherical/ellipsoidal/geoidal) geographic coordinates.
(3) We have changed the scale, however, by expressing coordinates in millidegrees instead of degrees.
Since a map projection involves also doing (1) and (2) in addition to (3), which we have not done a projection. In particular, we have written geographic coordinates in degrees as geographic coordinates in millidegrees and then converted the number to base 30. This is not a projection.
Suppose you give me a map, how about the Carbury Creek Quadrangle, Oregon, Provisional Edition 1983. Now, if I cross out the scale in kilometers and draw one in in meters, and then cross out all the numbers on the map and write their base 30 (or hexadecimal, binary, or any other base) equivalents, have I changed the map projection into a new one? No, I haven't. The datum used to produce the map is still the same, and nothing I have done requires that the map be redrawn on its plane of projection. Likewise, the method described in Microsoft's patent has also not created a new projection.
Now, prove to me that they have created a new projection. (Since I have effectively disproven it.)
"Alcohol, Tobacco, Firearms, and Explosives" should be a convenience store, not a government agency.
...obvious and trivial patents is way more than enough to bankrupt just about any small or medium developer out there who gets accused of a patent violation by Microsofts teams of lawyers.(or some other large concern to be fair)
It doesn't matter if it's obviously bogus on a casual review, the fact remains that they have the most money to burn, and can drag almost anyone through the "justice" system until they are quivering whimpering paupers. It wouldn't take too many public and loud examples of this nature to throw quite the chill into independent development of the overt and above board kind. This is a company that successfully fought off the US government and pretty much got away with it. Their technical legal "loss" wasn't a loss,not to them anyway, when they are allowed to print up their own "money" to use to pay the fines, and that "money" went to brainwash the next generation into still using their products. How many other folks get to do that in this justice system? The only other example I can think of readily and off hand is the music industry folks getting busted for price fixing, and then forced to distribute copies of their songs as the fine. Same deal, IP, trivially cheap to mass produce, being used as money to pay off a court fine. Sweet deal for them, and it's not a loss, and not even close to being any inducement for them to "change their tune" on anything really. From one day to the next day, business as usual..
Then there's the question and observation that a construct such as "linux" isn't really owned by anyone, except the copyright Linus owns of course for the kernel. The groupings of softwares combined make a distro, various companies and orgs of both profit and not for profit status distribute these, along with individual files or groups of files called applications. Once the patent wars begin, which they will eventually (it is still in the scouting and skirmishing phase now, along with accumulating patent-ammunition stockpiles to use then), you'll see most of those dry up, be made illegal to distribute,be forced to either pay just scads of bewildering royalty licenses because of these thousands of trivial but still legal patents, or be forced into this dubious ephemeral "underground" existence at least to the point of a similarity to what the P2P networks are enjoying now,complete with take downs, raids, lawsuits, actual loss after loss in the court system, now even affecting joe and jane small fry when they choose to do it, and increasingly they are choosing to do it.
It is quite conceivable that software patenting per se will result in just a small number of "legal" (I am only speaking inside the US for now), operating systems and application vendors in the not too far away future.
They are not accumulating all these patents as a hobby.
It's an application for patent. It's not a patent yet, and hopefully it won't make it through the gauntlet. That said, it could easily make it through the gaps in the system.
Don't get bogged down by the "base 30" part. Patent applications have a "preferred embodiment" or current invention part, but this is not the part you have to worry about. Patent applications, if you have a good patent lawyer, try to cover off as broad a space as possible without getting summarily struck down. Look at the claims; they're the part that other companies can run afoul of. Other parts are supporting documentation to show that yes, it is an invention (you must make mention of the device the software runs on in software patents, for example) and to preemptively strike down the examiner's questions.
So, what we have here is a patent on turning lat/long information into fixed point (trivial), then represented as base anything. It does not have to be in a URL. It does not have to be base 30.
I don't think this one should stand.
I'm wondering how many other software developers hang out on here, and what they think of software patents. I'll say, for my part, that I've never had to refer to patents to help me in my line of work in any way whatsoever. Never mind the triple-indemnity-if-you-knew clauses. If you're given a problem to solve, you cover them with assumptions to make the problem easier, standard techniques and analysis, and other peoples' components to bring up the shortfall. Patents that are broad enough to worry about rarely contain content that's helpful.
I can't foresee patents helping software developers - unless you count learning to dance in minefields... 'helpful' :)
I don't even actually see Microsoft being the main worry on an ongoing basis. I see our industry being held to blackmail by IP "holding companies" who do not develop software, and thus who cannot be threatened or counterattacked in the same way as Microsoft can.
-- Ritchie
Binary geeks can count to 1,023 on their fingers
I love base-30 systems - we have one that is about to expire, and has already caused all sorts or predictions of doom and gloom here on slashdot: The Serial Number for the Vehicle Identification Numbering system (VIN).
Of course that was created to make it harder for your local sherrif and local auto dealer from writing down the VIN as "I2FO0000" instead of "12E00000" . . . despite the reality that rolling over the VINs at the end of this decade should only be an inconvenience for the corporate executives that have the "I must start numbering everything at 1" mind-set. But getting back on topic . . .
Let's break this down:
Sigh. Nevermind. Our entire society is built upon a foundation of information sharing - the public school system. We spend 12 years forcing our children to accept and regurgitate information from a small number of sources (teachers), and to share information freely amongst themselves (recitation in class), and then to prove that they know it (tests) . . . only to turn around and slap the label "Intellectual Property" and to forbid them from doing precisely what we just spent 12 years training them to do.
Self-defeating.
Information is the INFRASTRUCTURE of technological advance. Just like roads and the electrical grid is the infrastructure of modern society. Any nation that figures this out, and acts to build its infrastructure will do precisely what the United States did between 1865 (backwater nation fighting a civil war) and 1945 (superpower) . . .
Chivalry is not dead, it's just frequently misspelt. - M. Langley