Thanks for replying - it's good to hear from someone who knows. Why not create an account?
What are you doing now? Are you a patent attorney or patent agent?
I have a lot of questions for an ex-examiner. For example, most current software patents have all the references starred - "cited by examiner". Does that mean the applicant did no prior art search?
They aren't claiming wildcards in general, nor even wildcards in DNS. They're claiming DNS wildcards used to host multiple user's domains on one virtual IP address.
Stallman is at least half right there, but I have a feeling he doesn't quite understand the ideas behind claims drafting. The applicant doesn't want a long, complicated claim - he wants the shortest, simplest, and therefore broadest claim possible. The examiner forces him to add elements to his independent claim (claim 1 here) to narrow it. Because they both know that in a few years people will be previewing music in a similar, but different way, and the applicant would like to own that too.
And it's not a patent on subdomains. Given that rather basic misunderstanding, isn't it possible that the guy who examines patents all day is right, and the slashdot crowd is wrong?
If this method is truly novel, it could be patentable. It wouldn't be a patent on "9, and any number divisible by 9" but rather a Method for Detecting Transposition Errors in Arithmetic.
Of course slashdot would say that you had patented arithmetic, period.
When the whole world has adopted the system, and when they cannot prove with absolute certainty that they were the first to come up with the idea, their idea will be thrown out.
Wrong on both counts. That fact that there are many infringers does not impair the validity of a patent. In fact, in one respect it strengthens the validity - nobody can claim that the patent has no practical utility, which is one possible challenge to a patent.
Once a patent has been issued, it is presumed valid. In other words, the patent holder is not required to "prove with absolute certainty" that he invented his invention - rather the burden is on infringers to show that someone else invented the invention, or find another means of invalidating the patent.
They aren't moronic. They do their job pretty well, within the rules. It's not their fault that slashdotters don't understand the rules and don't bother reading patents before attacking them. The patent appears to be 6,687,746.
It's a hard patent to read, but the key claim is the use of a DNS wildcard entry to handle user's subdomains. The applicants claim that as of August 1999 everyone was entering separate DNS records for each subdomain.
Can you find prior art? A published description of using a single DNS wildcard for user's subdomains prior to 8/99?
If so, great. You can kill this patent. But how should the examiners have been aware of this? Look at the large number of references the examiners plowed through - where do you expect them to draw the line?
Most of the "glaring mistakes" that slashdotters attribute to the PTO are simply myths. Lazy people who couldn't take the time to read a dense, techno-legal document made up alarmist stories based on the patent title or a few words that caught their eye. And you're lapping it up, like a right-winger inhaling his Rush Limbaugh horror stories or a left-winger gobbling up his Al Franken product.
That was well written, but it gave me a thought. Have we become so desensitized that exposing "gross wrongdoing" no longer matters? If Nixon were president today, would anyone really care about Watergate?
Clinton lied, and it was not the innocent action his supporters portray. He was willing for Monica Lewinsky to go through life branded as a liar who made up her affair with the president, and he lied to cover up a record of harrassment. It's becoming pretty clear that Bush lied about the weapons of mass destruction. Kerry apparently lied about his Vietnam experience and introduced fake vietnam vets in front of congress to help end the war. Reagan was largely out of it.
We've got so used to the Dems and Republicans flinging mud at each other that I don't think we notice that accusations any more.
XFCE isn't the only lightweight WM. There are more lightweight WM's than pigs. Blackbox, fluxbox, IceWM, etc. I used to run AfterStep on a 233Mhz machine and it was fine.
Well I don't know if Cobind really adds anything, but there is a real need for a distro that takes responsibility for the user experience and delivers a seamless, resilient, newbie-friendly product. It will take substantial effort, and it wont' appeal to either Unix geeks or Windows power users. Cobind is at least on the trail, with only one of each program.
I'm disgusted with the total lack of integration in most existing distros. For example, Red Hat's package of lynx tries to use xli to view images, but xli is not included. You can call that a bad example, because Lynx is a 'power user' tool, but it's typical. None of the Linux distros appears to have been QA'd.
What if the creator is neither cruel nor kind, as we understand them, but has a mind as far beyond ours as ours is beyond the primitive guidance system of an amoeba? Aren't our notions of cruelty and kindness deeply rooted in our reality as transient, mortal animals? Put differently, we don't apply these ideas to bacteria because they exist on such a different plane of development.
"Google is a nice little search engine, but nothing compared to my vision."
That's the eternal problem of comparing realities to visions. The realities always look pretty dull in comparison. No real government can measure up to the visions of Communism or Libertarianism.
But maybe it is a bit ironic after all. The core of the document was textual. Was this argument enhanced by being presented in a tiny font in a 3-column format? Why didn't they just put up a web page?
In commissioning this "designed" document, Microsoft seems to ignore the wisdom of Microsoft users. Word documents are ugly, and they shape the world. Fancy design doesn't add to an argument.
But if I read the cited metadata correctly, the document was created in desktop publishing program Quark. Now it's true that Microsoft doesn't make a direct competitor to Quark - nevertheless it's somewhat ironic that they didn't eat their own dogfood.
After all, if the Office suite is not adequate for communicating business ideas, like the superiority of one's product, why should anyone buy it?
Assuming that a language has no major design mistakes, there is generally a tradeoff between simplicity of the language and simplicity for the programmer.
Lisp is beautifully, terribly simple, but it may not be simple to write a useful program in Lisp.
C is "simple" with its mere 31 keywords, but the flip side is typing if(!strncmp(input_name, name, strlen(name))) when you really mean "if input_name = name".
Perl and Python are simpler for the programmer, at the cost of great complexity under the covers.
AutoCAD isn't written in Lisp - it uses Lisp as a scripting language. And the implementation is broken (don't remember the details) so that real Lisp idioms won't work. AutoLISP scripts are more like Visual Basic recast into prefix notation. Not heavy on recursion and small functions.
Anyhow, Autodesk has de-emphasized AutoLISP in favor of Visual Basic, which they also support. This is all secondhand - I haven't touched this stuff in years.
Do we need yet another impractical language? Look, Perl started off solving problems in the real world and was then developed into a "real" language. These pie-in-the-sky types all have the same problem, and Larry Wall nailed it. They're inventing a language around some theoretical idea, rather than looking at what works in the real world and gently enhancing it.
Even if the main idea is a great one, the language always sucks because it is too much enslaved to that one idea.
And the funniest part is that the language of the week is always on a web site served up by C/C++ and (Linux|Windows).
I'm looking for insight from people who've used Frame recently. Does it still have a substantial lead over Word for handling large docs? People have written books in Word. How about free software like Lyx, Kword, etc - what does Frame offer over them?
Has Frame improved or stagnated under Adobe's stewardship?
I found your comment interesting, and would like to see amplification of most of the points. I sense that you are coming from a graphic design viewpoint, rather than a "document processing" viewpoint. Could any markup language ever satisfy you, or is it too counterintuitive to write markup?
Can you explain more about modern authoring workflows? Are there open standards? What products handle these workflows well? LaTex is mostly used to create black and white technical papers, a job it does pretty well. If a better markup processor could handle more graphically ambitious goals, that would be interesting. But I suspect graphics professionals would stick to a GUI no matter what. I suggest you create an account so your comments will be more visible.
That's an interesting angle, but Frame is a heavy duty document processor, not really comparable to a word processor although Word is catching up. So I don't think Frame users pick the OS first, and then pick Frame as a generic word processor. Rather, I think a company sets up a tech writer with a Frame workstation and has to decide the underlying OS based on what they're comfortable with.
The thing that confuses me is that now that Macs are BSD based, shouldn't it be relatively simple to port the Solaris version to MacOSX?
Not at all; the difficulty is not in the POSIX bits - read/write/open/close - but in the GUI. A well behaved Mac app needs to use unique Apple API's correctly, such as Cocoa. Besides, support can be a bigger issue than initial porting. I know of products that could be ported to Linux in a heartbeat, except that the support issues scare the owners.
Anyhow, Frame is essentially a corporate product and corporations have not accepted the Mac to any great extent. It's used in graphic arts, prepress, etc. but most IT departments would rather avoid them. The Mac mostly sells to consumers and independent professionals.
Slashdot Mirror
Thanks for replying - it's good to hear from someone who knows. Why not create an account?
What are you doing now? Are you a patent attorney or patent agent?
I have a lot of questions for an ex-examiner. For example, most current software patents have all the references starred - "cited by examiner". Does that mean the applicant did no prior art search?
Interesting, thanks for the link.
They aren't claiming wildcards in general, nor even wildcards in DNS. They're claiming DNS wildcards used to host multiple user's domains on one virtual IP address.
Stallman is at least half right there, but I have a feeling he doesn't quite understand the ideas behind claims drafting. The applicant doesn't want a long, complicated claim - he wants the shortest, simplest, and therefore broadest claim possible. The examiner forces him to add elements to his independent claim (claim 1 here) to narrow it. Because they both know that in a few years people will be previewing music in a similar, but different way, and the applicant would like to own that too.
Actually, examiners work in pretty narrow areas. Check out other patents allowed by the same examiner. They're mostly computer/internet patents.
And it's not a patent on subdomains. Given that rather basic misunderstanding, isn't it possible that the guy who examines patents all day is right, and the slashdot crowd is wrong?
If this method is truly novel, it could be patentable. It wouldn't be a patent on "9, and any number divisible by 9" but rather a Method for Detecting Transposition Errors in Arithmetic.
Of course slashdot would say that you had patented arithmetic, period.
Wrong on both counts. That fact that there are many infringers does not impair the validity of a patent. In fact, in one respect it strengthens the validity - nobody can claim that the patent has no practical utility, which is one possible challenge to a patent.
Once a patent has been issued, it is presumed valid. In other words, the patent holder is not required to "prove with absolute certainty" that he invented his invention - rather the burden is on infringers to show that someone else invented the invention, or find another means of invalidating the patent.
They aren't moronic. They do their job pretty well, within the rules. It's not their fault that slashdotters don't understand the rules and don't bother reading patents before attacking them. The patent appears to be 6,687,746.
It's a hard patent to read, but the key claim is the use of a DNS wildcard entry to handle user's subdomains. The applicants claim that as of August 1999 everyone was entering separate DNS records for each subdomain.
Can you find prior art? A published description of using a single DNS wildcard for user's subdomains prior to 8/99?
If so, great. You can kill this patent. But how should the examiners have been aware of this? Look at the large number of references the examiners plowed through - where do you expect them to draw the line?
Most of the "glaring mistakes" that slashdotters attribute to the PTO are simply myths. Lazy people who couldn't take the time to read a dense, techno-legal document made up alarmist stories based on the patent title or a few words that caught their eye. And you're lapping it up, like a right-winger inhaling his Rush Limbaugh horror stories or a left-winger gobbling up his Al Franken product.
You mean that given the choice of Linux distros with too many choices and distros with too few choices, you choose to avoid the choice altogether?
That was well written, but it gave me a thought. Have we become so desensitized that exposing "gross wrongdoing" no longer matters? If Nixon were president today, would anyone really care about Watergate?
Clinton lied, and it was not the innocent action his supporters portray. He was willing for Monica Lewinsky to go through life branded as a liar who made up her affair with the president, and he lied to cover up a record of harrassment. It's becoming pretty clear that Bush lied about the weapons of mass destruction. Kerry apparently lied about his Vietnam experience and introduced fake vietnam vets in front of congress to help end the war. Reagan was largely out of it.
We've got so used to the Dems and Republicans flinging mud at each other that I don't think we notice that accusations any more.
XFCE isn't the only lightweight WM. There are more lightweight WM's than pigs. Blackbox, fluxbox, IceWM, etc. I used to run AfterStep on a 233Mhz machine and it was fine.
Well I don't know if Cobind really adds anything, but there is a real need for a distro that takes responsibility for the user experience and delivers a seamless, resilient, newbie-friendly product. It will take substantial effort, and it wont' appeal to either Unix geeks or Windows power users. Cobind is at least on the trail, with only one of each program.
I'm disgusted with the total lack of integration in most existing distros. For example, Red Hat's package of lynx tries to use xli to view images, but xli is not included. You can call that a bad example, because Lynx is a 'power user' tool, but it's typical. None of the Linux distros appears to have been QA'd.
What if the creator is neither cruel nor kind, as we understand them, but has a mind as far beyond ours as ours is beyond the primitive guidance system of an amoeba? Aren't our notions of cruelty and kindness deeply rooted in our reality as transient, mortal animals? Put differently, we don't apply these ideas to bacteria because they exist on such a different plane of development.
That's the eternal problem of comparing realities to visions. The realities always look pretty dull in comparison. No real government can measure up to the visions of Communism or Libertarianism.
Oh, absolutely. Dreadful language. But then, "There are two kinds of languages. The kind that suck, and the kind that nobody uses."
But maybe it is a bit ironic after all. The core of the document was textual. Was this argument enhanced by being presented in a tiny font in a 3-column format? Why didn't they just put up a web page?
In commissioning this "designed" document, Microsoft seems to ignore the wisdom of Microsoft users. Word documents are ugly, and they shape the world. Fancy design doesn't add to an argument.
But if I read the cited metadata correctly, the document was created in desktop publishing program Quark. Now it's true that Microsoft doesn't make a direct competitor to Quark - nevertheless it's somewhat ironic that they didn't eat their own dogfood.
After all, if the Office suite is not adequate for communicating business ideas, like the superiority of one's product, why should anyone buy it?
Assuming that a language has no major design mistakes, there is generally a tradeoff between simplicity of the language and simplicity for the programmer.
Lisp is beautifully, terribly simple, but it may not be simple to write a useful program in Lisp.
C is "simple" with its mere 31 keywords, but the flip side is typing if(!strncmp(input_name, name, strlen(name))) when you really mean "if input_name = name".
Perl and Python are simpler for the programmer, at the cost of great complexity under the covers.
AutoCAD isn't written in Lisp - it uses Lisp as a scripting language. And the implementation is broken (don't remember the details) so that real Lisp idioms won't work. AutoLISP scripts are more like Visual Basic recast into prefix notation. Not heavy on recursion and small functions.
Anyhow, Autodesk has de-emphasized AutoLISP in favor of Visual Basic, which they also support. This is all secondhand - I haven't touched this stuff in years.
Negative comments about a new language are whining.
Negative comments about the negative comments are insightful.
Do we need yet another impractical language? Look, Perl started off solving problems in the real world and was then developed into a "real" language. These pie-in-the-sky types all have the same problem, and Larry Wall nailed it. They're inventing a language around some theoretical idea, rather than looking at what works in the real world and gently enhancing it.
Even if the main idea is a great one, the language always sucks because it is too much enslaved to that one idea.
And the funniest part is that the language of the week is always on a web site served up by C/C++ and (Linux|Windows).
Well, you really showed them. This'll teach them to patronize women. Oh wait.
I'm looking for insight from people who've used Frame recently. Does it still have a substantial lead over Word for handling large docs? People have written books in Word. How about free software like Lyx, Kword, etc - what does Frame offer over them?
Has Frame improved or stagnated under Adobe's stewardship?
I found your comment interesting, and would like to see amplification of most of the points. I sense that you are coming from a graphic design viewpoint, rather than a "document processing" viewpoint. Could any markup language ever satisfy you, or is it too counterintuitive to write markup?
Can you explain more about modern authoring workflows? Are there open standards? What products handle these workflows well?
LaTex is mostly used to create black and white technical papers, a job it does pretty well. If a better markup processor could handle more graphically ambitious goals, that would be interesting. But I suspect graphics professionals would stick to a GUI no matter what.
I suggest you create an account so your comments will be more visible.
Not at all; the difficulty is not in the POSIX bits - read/write/open/close - but in the GUI. A well behaved Mac app needs to use unique Apple API's correctly, such as Cocoa. Besides, support can be a bigger issue than initial porting. I know of products that could be ported to Linux in a heartbeat, except that the support issues scare the owners.
Anyhow, Frame is essentially a corporate product and corporations have not accepted the Mac to any great extent. It's used in graphic arts, prepress, etc. but most IT departments would rather avoid them. The Mac mostly sells to consumers and independent professionals.