> Yes but priv sep requires code changes for each privilege seperated application. Some apps are particularly difficult to do this with, like ssh. But after a lot of effort, they did it. So there is a huge gap between "can be" and "is".
The openssh team did a good job there, but openssh is far from OpenBSD exclusive. Privilege seperation is available for at least each unix like platform on which openssh runs.
There is quite some overlap between the OpenBSD and OpenSSH teams, but as said, this is nowhere OpenBSD exclusive, and for example on FreeBSD and many Linux distributions you will get the exact same privilege seperation. In other words, this is a good reason to use openssh, but in itself not an argument for (or against) OpenBSD.
> As I've said before, OpenBSD is not perfect (what is?). But it is still one of the best choices for security and continues to advance. It obviously is not going to fit into every need.
It is a very good choice, I agree there. I just wanted to point out that privilege seperatiopn is not an operating system issue, it is an application issue, and for many relevant applications where OpenBSD has a default install with privilege seperation, you will find that other platforms have the same.
If they indeed talk to their customers and try to get the trojans removed, then this may be a good idea.
I find it kindof funny however that problems with their nameservers is what finally got them to act, while they can quite prevent such infected PCs from messing up for their other customers.
A while ago I wrote a bit about preventing flooding of a nameserver that with a bit of tuning would quite help to prevent the slowness of their nameservers regardless of those trojans. What is more, it would make trojaned PCs that flood the nameservers mostly unusable without hindering normal clients, giving their customers more of an incentive to deal with it themselves.
> Setting up spamd with dspam or you other filter of choice is handy cause what gets through spamd usually won't get through the other.
On FreeBSD:
cd/usr/ports/mail/spamd && make install clean && cd../dspam && make install clean
I'm pretty sure equivalent packages are in pkgsrc on NetBSD.
You may need to edit the config files for both in/usr/local/etc/ and uncomment a line here and there and change a few lines to match your config and you are done.
> While I cannot say it for NetBSD or FreeBSD, Linux isn't documented so much as HOWTOed. I don't find a good man page very often for something done with Linux.
Definitely agree there. For a quick setting up of things that is all that is needed tho. For understanding things I rather prefer any of the BSDs and their documentation.
> You don't tweak the kernel because the default one has almost everything that is supported. It makes the kernel bigger than it might otherwise need to be, but if you've got more than 16 mb of memory it doesn't matter.
If you are security consious, you should always be trying to only have those things installed and in memory that you need. Less code == les vulnerabilities, that even holds on OpenBSD.
> What does this have to do with FreeBSD? Although FreeBSD is the only BSD with support for DRI accelerated OpenGL, X.org/XFree provides the same 2D graphics hardware support on all platforms.
Of course there is also the little thing of there beign an official nvidia driver for FreeBSD and not for the other ones.
ANd yeah, this 'only' matters for OpenGL, so who cares?
Well, everyone running a modern desktop system basicly since at least KDE and Gnome based desktops become quite a bit more responsive with hardware opengl support and your overal performance when making some what heavier use of your desktop will be better.
But... aren't we getting rather offtopic here? (looks at article and sees something about a new Dragonfly release)
privilege seperation is a technique that can be employed on any Unix and is nowhere OpenBSD specific, their developers do use it where they can tho it seems.
> OpenBSD is the suggested often for a router because it is a tight little system with a fair number of security enhancements
Definitely but it is still lacking FreeBSD style jails (not often an issue on a router but can be very nice for seperating your mail environment a bit better from the router itself, see below)
> and it's pf packet filter is native to OpenBSD, thus more tightly integrated and tuned for Open.
Its native for open, but integration in freebsd seems to be pretty good. What is more, FreeBSD gives a choice of 3 different packet filters
> It's partly that you don't want a router cracked and partly cause you want the best packet handler, pf is that....
> Plus you can set it up as your mail relay and stop spam, and yadda, yadda, yadda... It's a generally nice small system and the ports with it almost all run without fuss.
Uh, I can do the same with virtually any Unix, and setting this up with NetBSD, FreeBSD or Linux is really easy and well documented.
When you want your router to also handle your mail, I'd prefer a system where I can run the smtp services in a well seperated environment, which means more then a mere chroot. Despite all the good work of Wietse, smtp, esp together with virus and spam checking and blocking is a too complex thing to trust it running on my router without such seperation. (perl, substantial amount of perl scripts for amavis and spamassasin etc)
Regardless, I seconf your opinion that openbsd would be the better choice, but mostly because of the first reason, their very nice security enhancements.
> I know there is a lot of word processing software that goes way back. That's not the point - we're not trying to combine word processing with a computer.
Not as such, we are ifaik looking for a variety of techniques and nethods that existed at the time of the patent and can simply be combined to create what is described in the patent.
> We're trying to find something along the lines of a WYSIWYG interface, because then it would be easier to show that it is obvious to implement a paper form (with a mandatory information mark) on a computer screen. (And let me point out that I didn't choose to go this direction. Were I to build an argument that this is obvious, I would have used an existing computer interface, complete with mandatory user-input at certain stages, and brought in something like curses, ncurses, or something teaching and showing the advantages of a GUI.)
Are we looking for prior art that does exactly what is claimed or are we looking for its individual components and show why it does not take any hard thought to combine them to what is claimed?
Both from the text you refered me to earlier, and from what you say here, I get the feeling that prior art and obviousness can only be proven at the same time, in which case I wonder, why are new (as in not pre-existing) and non-obviousness seperate requirements if you can basicly only disprove the later if you can disprove the first?
> Well, for a long time, a computer's user interface was wires and blinking lights. For awhile, it was push button. For awhile, it was punch cards. The command line interface still persists to this day and it doesn't come close to any real world metaphor that I'm aware of. I am a linux geek, but I wouldn't have a clue where to start looking for a honest-to-God document metaphor before 1990 or 1985.
Hmm, sorry again for a double answer, but I grew up with machines like the CBM PET and the C64.
On both machines I used word processing software (visiwrite on the C64) that had a document inerface, which predates 1985, and even 1980.
Rank Xerox had a machine in the late 70s which used a graphical user interface and had a document metafor for many things. In 1990 I was working at IBM and was using lots of software on their mainframes that predate 1980 that used a docoment metafor and even a form metafor.
There is really a lot predating the line you put there.
A document metafor can be implemented using a text only interface, it in no way requires a gui. It does require a full screen interface, but even on Unix those existed way before 1980.
> That and the next three sections explain the three basic components necessary to defeat a patent claim as obvious. As soon as you stray from what the references teach and into "the knowledge of a person of ordinary skill in the art" you're pretty much bluffing (because the attorney can find an expert who will swear in an affidavit that he wouldn't know how to do it, and unless you (the patent examiner) likes working free overtime at the office, there really isn't anything you can do about that argument (this is one of the situations where you pretty much have to issue the patent and let it get straightened out in court - you're simply not paid to deal with that situation.))
I understand that, but I think that is also where part of the problem comes from.
As inventor, I do not have the time and resources to fight such things if I am to do usefull iinventions.
In the end, the claim that many make is that the patent system at least in case of software, defeats its own purpose.
This is why I and I think many with me, believe that sopftware patents are generally a bad idea, even if it is possible to come up with some valid software patents.
Again, thanks for clearing up quite a few things. I will later try to make a more complete post about that specific MS patent.
> I don't expect you to understand all of the law, but I think everyone would be wise to recognize the limits of their understanding. It's far better to ask, "Why wasn't this declared obvious?" than "Those idiots didn't know this is obvious?" I do my best to follow this advice as well. I don't mean to single you out as the guy calling people names, but rather I was calling attention to how complicated and misunderstood the concept of obviousness in patents was, and you took the bait.;)
I think that that was exactly what I was asking while also providing an indication as to why this did seem extremely obvious to someone skilled in the art (but not being a lawyer)
Assembled from multiple sources in this seems to rather be about a legal standard for proof, and not about the actual definition of this concept called obvious. It is the job of a patent attorny to translate this into legally correct proof, and imho also the job of a patent examiner to investigate when a claim of technical obviousness is made and documented. Patents are about inventions, legal definitions are merely a means of correctly writing them, no more and no less.
That someone claiming something is not legal proof I understand, anyone can claim anything they like.
I was however not merely claiming that anyone could come up with this, but that it is a logical application of known ideas and concepts, and indeed that it is extremely likely that someone skilled in the art would come up with the exact same or at least with something that is conceptually identical.
Just me claiming that is not enough for legal proof, but this claim should be more then enough for a patent examiner to investigate this.
Again, the purpose of patents is to promote usefull inventions, not to provide job security for lawyers.
Sorry for the double reply, but I thought I had to add this little bit.
> would then explain how the spell checking feature could be combined with the email client, for example as a component of the software that takes the email body as input and checks the spelling as it would a word processor document. Then, and only then, have I built a prima facie case of obviousness. (I've even skipped a few parts of the Graham factual inquiries for a proper obviousness rejection.)
Well, I do not know the details but understand the idea you are getting at.
The problem here is wording, and not the actual content of what is being said.
Following your example, I could explain how marking is used on a paper form, how a computer user interface asking for user input is merely an electronic equivalent of a form, it is presenting the user with text questions on which answers are expected. Then I could explain how on a traditional form highlighting is used, and how the exact same thing can be done regardless of the form being on paper or another display. I could even explain that paper or a display make no difference for the concept of a form, it is a medium, and nothing more.
I realize that actual proof in court requires more exact language then that, but honestly, if you are discussing this with a non lawyer, you have to look for the meaning of what a person is trying to say, and not at the exact use of legal language. I am sure you can find a lawyer who can put what I just said in official legal language and prove the case.
> However true that may be, it is nonetheless hyperbole and meaningless with regard to the obviousness of the invention. (I wouldn't even agree that it's true without qualification - certainly someone could come up with an equivalent function by some other mechanism.) This does play into the question, "Is this a good implementation?" It has nothing to do with whether or not it was obvious. Obvious under 35 USC 103 means, generally speaking, "the concept was assembled from more than one source," not the dictionary definition of "easy to understand".
Obvious in the dictionary definition can as well mean that it is merely a logical consequence of what is already known. Easy to understand is something that can only really be said after the fact, ie, when you know the exact concept. Obvious as I used it is about the fact that an invention is a logical consequence of already existing inventions, and is something that a person skilled in the art is extremely likely to come up with when having to solve the problem at hand.
Regardless, you can try discussiing the legal standard that proof has to comply with, and I am interested in that, but when looking at an invention as a technical person skilled in the art, the legal definition is not very relevant.
Why not? well, read back and see why patents exist. They are there to promote usefull inventions, not to keep lawyers employed. If the legal definition is so far away from what people in the field use that it becomes useless in the field then the legal definition is broken imho.
> I never had any reason to order the prosecution record, of course, so it might have had nothing to do with the issue of obviousness. It could have been that IBM's attorney's argued that THEIR system was ONLY for use on their corporate intranet, a use for which there was no prior art whatsoever in 1990. Maybe they argued that their system was only for use on 256 color displays, using a shading system that was not available before, and therefore no reasonable argument of obviousness would hold up on appeal. Who knows? You'd have to investigate more than the patent to find out how it got issued.
Intranet is a meaningless word, it is just a network. WHatever borders you put around your network dont make any difference for the network from a technical point of view.
Number of colors? pseudo color might have been an invention, indexed colors might have been an invention, using a color to indicate a specific status of an item or field? please..
> None of your are even remotely suggests a computer system.
So the sole fact that something involves a computer system makes it new and non obvious?
> This has absolutely no basis whatsoever in the law. Please see MPEP 2141-2144 to understand what the term "obvious" means as used in 35 USC 103. You are using the dictionary meaning of the term "obvious" which has basically no bearing on 35 USC 103.
As a layman I am supposed to be aware of the law (ignorance is no excuse). If the law uses a more exact definition of a word then the dictionary definition that is fine and often required. If the law uses a definition that is totally unlike the common or dictionary definition then you simply cannot expect people to even remotely understand the law. This is twisting words in order to cloud things.
Regardless, the idea is not new in any way and it is basicly the first thing that anyone skilled in the art would think of when having to solve the issue at hand. Explain to me how this is non obvious.
Marking fields on a form that are mandatory has been common practise for a very long time, most likely for as long as the concept of a form exists, and at any rate longer then the USA exists as a nation.
Highlighting something as a means of indicating a special status has been in use for about as long as forms of writing exist.
This all should be considered prior art as meant by 102.
103 starts with the following:
(a) A patent may not be obtained though the invention is not identically disclosed or described as set forth in section 102 of this title, if the differences between the subject matter sought to be patented and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art to which said subject matter pertains. Patentability shall not be negatived by the manner in which the invention was made.
The fact that the form is displayed on a screen instead of on a piece of paper does not make it non obvious, it is a matter of applying a known idea in a very obvious way that is functionally and conceptually identical to the prior art.
Since you seem to be well informed and actually involved in this process, you might be able to answer some questions here:
> Did you know that IBM held a patent for highlighting an input field in a user interface to indicate to the user that the field is mandatory?
Could you explain why in the world that patent was granted? There is no check on something being obvious anymore? Or does the fact that something happens to involve a computer automatically make it new and non obvious?
Just to be utterly clear here, hilighting mandatory fields on a paper form has been used for longer then computers exist.
Could you explain how Microsoft managed to get a patent on using xor to make a cursor blink? (this is an obvious method for anyone who ever did a little bit of assembler coding at the very least)
Can you explain the famous Amazon one click patent and it not being dismissed inmediately because of being utterly obvious? (this seems to be the most valid one of the 3 mentioned in this post actually)
Can you point us at any software patents that actually comply with the requirement of providing enough information so that a practiser in the field the patent applies to can reimplement it based on the text of the patent?
Thanks for your corrections but it still makes no sense at all.
You can selectively enforce a patent that has been granted, publication of patent applications does not vhange anything there because granted patents have to be published anyway, and can indeed be enforced selectively or not at all without becomming invalid.
Take a little peek maybe at how a company like IBM has been using and is still using patents, they do exactly what you say is impossible now, and do it with patents that have been granted and as such are public.
WHat you cannot do is try to keep a patent in limbo and as such unpublished for an extended period of time, but the sheer amount of patents out there, and the fact that in many cases people explicitly do not look for patents (knowingly infringing one is way more expensive then unknowingly doing so) makes that publication helps little if at all.
A vigorously enforced patent may get enough publicity to no longer be able to claim you did not know about it (some of the mp3 related patents, the now expired ones relating to the gif format etc for example)
> That is why BSD license is not as good as the GPL.
Guess it never occured to you that this is actually desired by those using the BSD license?
If forcing open source on anyone who derives from your work is your purpose then the GPL is better (this is quite legitimate, but not what everyone wants)
If making available reference implementations of a standard for anyone to use regardless of OSS is your purpose, then the BSD license is definitely better. (note that even people in the FSF suggest this)
Hmm, have been out of the loop with regards to modern phones for a bit, but my now ancient Ericsson T39m comes with a ringtone editor built in, pc software for editing ringtones, and PalmOS software for editing ringtones.. did things change that dramatically in the last few years?
> My wife's old nokia phone didn't have perfect reception here in Seoul, but it was usually alright.
Hehe, well, Nokia usually sucks when it comes to decent reception.. but hrm, many people have them or have something rather comparable.
> CDMA works perfectly fine here though, I can use my phone on a moving subway (and my phone is an old Samsung). I know there's a much smaller area to cover than the States, but it seems that even in major cities in the States, people have some problems with coverage. Which seems to be pretty unreasonable in this day and age.
> So instead, they just put on board members that will do the least damage from their perspective.
Thats obviously what they are doing. It is also exactly why many consider the current US government to be a fascist government, they comply very well with the original definition of it (note that this is not the same as a nazi government, which they are definitely not, also note that I am in no way comparing Bush to Hitler here, I am pointing at the original definition of fascism, which is basicly putting industry in charge of government. It is not going to be good for the people at all however, such governments without exception end up turning a substantial part of their population into slaves effectively)
> Protecting privacy reduces the power of government. Why does anyone expect any government to protect their privacy?
Maybe because they are there to serve the people and not the other way around?
Maybe because it is in the spirit of the constitution which they promissed to serve and protect?
> And your bashing of Bush may have gotten you mod points, but the Democrats are no slouches in the corporate malfeasance department. Terry McAuliff was up to his neck in Global Crossing, and the first person Ken DeLay called when the shit hit the fan at Enron was the Clinton administration's Treasury Secretary.
Sure, but republicans are supposedly advocating a lean and mean government with limited power, I have yet to hear democrats claiming the same.
Besides, 2 wrongs don't make a right. That the clinton admninistration messed up is no excuse whatsoever for Bush to mess up.
Hmm, do places like the Egyptian desert, jungle of northern Thailand or the mountains between France and Spain count as 'middle of nowhere' ? I found I could make and receive calls in those places without trouble whatsoever.
If coverage would be bad in areas with a relatively large human population that would be really bad, and I am sure that is no longer the case in the USA either, but honestly, when I read the stories here, or listen to my American friends who have cellphones, it looks like reading about how things were overhere more then a decade ago with regards to coverage.
I am currently sitting inside at my girlfriends house (in Berlin, Germany), and I have full 5 bar reception and perfect sound quality, and it may drop to 3 bars, but there is no place within an hour travel from here where inside coverage wont be good enough for reliably using a cellphone (and that includes all subway stations (which have repeaters here) and trains and office buildings and so on that I have been to)
The nearest place that I found without superb inside coverage is some 2 hours away from here by high speed train, and there you will find that the type of phone makes a big difference. My old Ericsson (T39m) with its external antenna still has a usable signal there, many cheap phones with internal antenna will no longer work reliably there when inside.
Ah yeah, having 4 or more networks to chose from (I can roam on all networks here) helps a bit there of course but in my 'home' area where I cannot use roaming, coverage is even better then overhere.
SA can query blacklists (and does in its default config) but, it uses those as an indication. If maps or xbl blocks an address, but all other indicators are saying it is not spam, SA will not mark the mail as spam normally. Of course you can configure that differently if you feel like it.
At any rate, rbls should only be used as one of the possible indicators for spam, none of the rbls is perfect or even uptodate, and reöying on them as the main or even only indicator for spam is just a very good way to block legitimate mail, while the effect on stopping spam is not even close to 70% (at least on the mail servers that I run, which serve a few hundred users each)
XBL seems to be one of the more usefull ones among the rbls, rbls aiming purely at dynamic/home IPs seem to be utterly useless in practise.
Slashdot Mirror
> Yes but priv sep requires code changes for each privilege seperated application. Some apps are particularly difficult to do this with, like ssh. But after a lot of effort, they did it. So there is a huge gap between "can be" and "is".
The openssh team did a good job there, but openssh is far from OpenBSD exclusive. Privilege seperation is available for at least each unix like platform on which openssh runs.
There is quite some overlap between the OpenBSD and OpenSSH teams, but as said, this is nowhere OpenBSD exclusive, and for example on FreeBSD and many Linux distributions you will get the exact same privilege seperation. In other words, this is a good reason to use openssh, but in itself not an argument for (or against) OpenBSD.
> As I've said before, OpenBSD is not perfect (what is?). But it is still one of the best choices for security and continues to advance. It obviously is not going to fit into every need.
It is a very good choice, I agree there. I just wanted to point out that privilege seperatiopn is not an operating system issue, it is an application issue, and for many relevant applications where OpenBSD has a default install with privilege seperation, you will find that other platforms have the same.
If they indeed talk to their customers and try to get the trojans removed, then this may be a good idea.
I find it kindof funny however that problems with their nameservers is what finally got them to act, while they can quite prevent such infected PCs from messing up for their other customers.
A while ago I wrote a bit about preventing flooding of a nameserver that with a bit of tuning would quite help to prevent the slowness of their nameservers regardless of those trojans. What is more, it would make trojaned PCs that flood the nameservers mostly unusable without hindering normal clients, giving their customers more of an incentive to deal with it themselves.
> Setting up spamd with dspam or you other filter of choice is handy cause what gets through spamd usually won't get through the other.
/usr/ports/mail/spamd && make install clean && cd ../dspam && make install clean
/usr/local/etc/ and uncomment a line here and there and change a few lines to match your config and you are done.
On FreeBSD:
cd
I'm pretty sure equivalent packages are in pkgsrc on NetBSD.
You may need to edit the config files for both in
> While I cannot say it for NetBSD or FreeBSD, Linux isn't documented so much as HOWTOed. I don't find a good man page very often for something done with Linux.
Definitely agree there. For a quick setting up of things that is all that is needed tho. For understanding things I rather prefer any of the BSDs and their documentation.
> You don't tweak the kernel because the default one has almost everything that is supported. It makes the kernel bigger than it might otherwise need to be, but if you've got more than 16 mb of memory it doesn't matter.
If you are security consious, you should always be trying to only have those things installed and in memory that you need. Less code == les vulnerabilities, that even holds on OpenBSD.
> What does this have to do with FreeBSD? Although FreeBSD is the only BSD with support for DRI accelerated OpenGL, X.org/XFree provides the same 2D graphics hardware support on all platforms.
Of course there is also the little thing of there beign an official nvidia driver for FreeBSD and not for the other ones.
ANd yeah, this 'only' matters for OpenGL, so who cares?
Well, everyone running a modern desktop system basicly since at least KDE and Gnome based desktops become quite a bit more responsive with hardware opengl support and your overal performance when making some what heavier use of your desktop will be better.
But... aren't we getting rather offtopic here? (looks at article and sees something about a new Dragonfly release)
> W^X, Propolice, Stackghost, priv sep, priv revocation, etc etc etc...
How about MAC and ACLs?
privilege seperation is a technique that can be employed on any Unix and is nowhere OpenBSD specific, their developers do use it where they can tho it seems.
> OpenBSD is the suggested often for a router because it is a tight little system with a fair number of security enhancements
...
Definitely but it is still lacking FreeBSD style jails (not often an issue on a router but can be very nice for seperating your mail environment a bit better from the router itself, see below)
> and it's pf packet filter is native to OpenBSD, thus more tightly integrated and tuned for Open.
Its native for open, but integration in freebsd seems to be pretty good. What is more, FreeBSD gives a choice of 3 different packet filters
> It's partly that you don't want a router cracked and partly cause you want the best packet handler, pf is that.
> Plus you can set it up as your mail relay and stop spam, and yadda, yadda, yadda... It's a generally nice small system and the ports with it almost all run without fuss.
Uh, I can do the same with virtually any Unix, and setting this up with NetBSD, FreeBSD or Linux is really easy and well documented.
When you want your router to also handle your mail, I'd prefer a system where I can run the smtp services in a well seperated environment, which means more then a mere chroot. Despite all the good work of Wietse, smtp, esp together with virus and spam checking and blocking is a too complex thing to trust it running on my router without such seperation. (perl, substantial amount of perl scripts for amavis and spamassasin etc)
Regardless, I seconf your opinion that openbsd would be the better choice, but mostly because of the first reason, their very nice security enhancements.
> I know there is a lot of word processing software that goes way back. That's not the point - we're not trying to combine word processing with a computer.
Not as such, we are ifaik looking for a variety of techniques and nethods that existed at the time of the patent and can simply be combined to create what is described in the patent.
> We're trying to find something along the lines of a WYSIWYG interface, because then it would be easier to show that it is obvious to implement a paper form (with a mandatory information mark) on a computer screen. (And let me point out that I didn't choose to go this direction. Were I to build an argument that this is obvious, I would have used an existing computer interface, complete with mandatory user-input at certain stages, and brought in something like curses, ncurses, or something teaching and showing the advantages of a GUI.)
Are we looking for prior art that does exactly what is claimed or are we looking for its individual components and show why it does not take any hard thought to combine them to what is claimed?
Both from the text you refered me to earlier, and from what you say here, I get the feeling that prior art and obviousness can only be proven at the same time, in which case I wonder, why are new (as in not pre-existing) and non-obviousness seperate requirements if you can basicly only disprove the later if you can disprove the first?
> Well, for a long time, a computer's user interface was wires and blinking lights. For awhile, it was push button. For awhile, it was punch cards. The command line interface still persists to this day and it doesn't come close to any real world metaphor that I'm aware of. I am a linux geek, but I wouldn't have a clue where to start looking for a honest-to-God document metaphor before 1990 or 1985.
Hmm, sorry again for a double answer, but I grew up with machines like the CBM PET and the C64.
On both machines I used word processing software (visiwrite on the C64) that had a document inerface, which predates 1985, and even 1980.
Rank Xerox had a machine in the late 70s which used a graphical user interface and had a document metafor for many things. In 1990 I was working at IBM and was using lots of software on their mainframes that predate 1980 that used a docoment metafor and even a form metafor.
There is really a lot predating the line you put there.
A document metafor can be implemented using a text only interface, it in no way requires a gui. It does require a full screen interface, but even on Unix those existed way before 1980.
> That and the next three sections explain the three basic components necessary to defeat a patent claim as obvious. As soon as you stray from what the references teach and into "the knowledge of a person of ordinary skill in the art" you're pretty much bluffing (because the attorney can find an expert who will swear in an affidavit that he wouldn't know how to do it, and unless you (the patent examiner) likes working free overtime at the office, there really isn't anything you can do about that argument (this is one of the situations where you pretty much have to issue the patent and let it get straightened out in court - you're simply not paid to deal with that situation.))
I understand that, but I think that is also where part of the problem comes from.
As inventor, I do not have the time and resources to fight such things if I am to do usefull iinventions.
In the end, the claim that many make is that the patent system at least in case of software, defeats its own purpose.
This is why I and I think many with me, believe that sopftware patents are generally a bad idea, even if it is possible to come up with some valid software patents.
Again, thanks for clearing up quite a few things. I will later try to make a more complete post about that specific MS patent.
Hmm, forgot to login there.
Oh, and on another note,
;)
> I don't expect you to understand all of the law, but I think everyone would be wise to recognize the limits of their understanding. It's far better to ask, "Why wasn't this declared obvious?" than "Those idiots didn't know this is obvious?" I do my best to follow this advice as well. I don't mean to single you out as the guy calling people names, but rather I was calling attention to how complicated and misunderstood the concept of obviousness in patents was, and you took the bait.
I think that that was exactly what I was asking while also providing an indication as to why this did seem extremely obvious to someone skilled in the art (but not being a lawyer)
Assembled from multiple sources in this seems to rather be about a legal standard for proof, and not about the actual definition of this concept called obvious. It is the job of a patent attorny to translate this into legally correct proof, and imho also the job of a patent examiner to investigate when a claim of technical obviousness is made and documented. Patents are about inventions, legal definitions are merely a means of correctly writing them, no more and no less.
That someone claiming something is not legal proof I understand, anyone can claim anything they like.
I was however not merely claiming that anyone could come up with this, but that it is a logical application of known ideas and concepts, and indeed that it is extremely likely that someone skilled in the art would come up with the exact same or at least with something that is conceptually identical.
Just me claiming that is not enough for legal proof, but this claim should be more then enough for a patent examiner to investigate this.
Again, the purpose of patents is to promote usefull inventions, not to provide job security for lawyers.
Sorry for the double reply, but I thought I had to add this little bit.
> would then explain how the spell checking feature could be combined with the email client, for example as a component of the software that takes the email body as input and checks the spelling as it would a word processor document. Then, and only then, have I built a prima facie case of obviousness. (I've even skipped a few parts of the Graham factual inquiries for a proper obviousness rejection.)
Well, I do not know the details but understand the idea you are getting at.
The problem here is wording, and not the actual content of what is being said.
Following your example, I could explain how marking is used on a paper form, how a computer user interface asking for user input is merely an electronic equivalent of a form, it is presenting the user with text questions on which answers are expected. Then I could explain how on a traditional form highlighting is used, and how the exact same thing can be done regardless of the form being on paper or another display. I could even explain that paper or a display make no difference for the concept of a form, it is a medium, and nothing more.
I realize that actual proof in court requires more exact language then that, but honestly, if you are discussing this with a non lawyer, you have to look for the meaning of what a person is trying to say, and not at the exact use of legal language. I am sure you can find a lawyer who can put what I just said in official legal language and prove the case.
> However true that may be, it is nonetheless hyperbole and meaningless with regard to the obviousness of the invention. (I wouldn't even agree that it's true without qualification - certainly someone could come up with an equivalent function by some other mechanism.) This does play into the question, "Is this a good implementation?" It has nothing to do with whether or not it was obvious. Obvious under 35 USC 103 means, generally speaking, "the concept was assembled from more than one source," not the dictionary definition of "easy to understand".
Obvious in the dictionary definition can as well mean that it is merely a logical consequence of what is already known. Easy to understand is something that can only really be said after the fact, ie, when you know the exact concept. Obvious as I used it is about the fact that an invention is a logical consequence of already existing inventions, and is something that a person skilled in the art is extremely likely to come up with when having to solve the problem at hand.
Regardless, you can try discussiing the legal standard that proof has to comply with, and I am interested in that, but when looking at an invention as a technical person skilled in the art, the legal definition is not very relevant.
Why not? well, read back and see why patents exist. They are there to promote usefull inventions, not to keep lawyers employed. If the legal definition is so far away from what people in the field use that it becomes useless in the field then the legal definition is broken imho.
> I never had any reason to order the prosecution record, of course, so it might have had nothing to do with the issue of obviousness. It could have been that IBM's attorney's argued that THEIR system was ONLY for use on their corporate intranet, a use for which there was no prior art whatsoever in 1990. Maybe they argued that their system was only for use on 256 color displays, using a shading system that was not available before, and therefore no reasonable argument of obviousness would hold up on appeal. Who knows? You'd have to investigate more than the patent to find out how it got issued.
Intranet is a meaningless word, it is just a network. WHatever borders you put around your network dont make any difference for the network from a technical point of view.
Number of colors? pseudo color might have been an invention, indexed colors might have been an invention, using a color to indicate a specific status of an item or field? please..
You see, the problem is that the
> None of your are even remotely suggests a computer system.
So the sole fact that something involves a computer system makes it new and non obvious?
> This has absolutely no basis whatsoever in the law. Please see MPEP 2141-2144 to understand what the term "obvious" means as used in 35 USC 103. You are using the dictionary meaning of the term "obvious" which has basically no bearing on 35 USC 103.
As a layman I am supposed to be aware of the law (ignorance is no excuse). If the law uses a more exact definition of a word then the dictionary definition that is fine and often required. If the law uses a definition that is totally unlike the common or dictionary definition then you simply cannot expect people to even remotely understand the law. This is twisting words in order to cloud things.
Regardless, the idea is not new in any way and it is basicly the first thing that anyone skilled in the art would think of when having to solve the issue at hand. Explain to me how this is non obvious.
Marking fields on a form that are mandatory has been common practise for a very long time, most likely for as long as the concept of a form exists, and at any rate longer then the USA exists as a nation.
Highlighting something as a means of indicating a special status has been in use for about as long as forms of writing exist.
This all should be considered prior art as meant by 102.
103 starts with the following:
(a) A patent may not be obtained though the invention is not identically disclosed or described as set forth in section 102 of this title, if the differences between the subject matter sought to be patented and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art to which said subject matter pertains. Patentability shall not be negatived by the manner in which the invention was made.
The fact that the form is displayed on a screen instead of on a piece of paper does not make it non obvious, it is a matter of applying a known idea in a very obvious way that is functionally and conceptually identical to the prior art.
> True, but only if you sold it, not gave it away.
That would still kill all GPL media players because such an additional restriction is not allowed under the GPL.
Since you seem to be well informed and actually involved in this process, you might be able to answer some questions here:
> Did you know that IBM held a patent for highlighting an input field in a user interface to indicate to the user that the field is mandatory?
Could you explain why in the world that patent was granted? There is no check on something being obvious anymore? Or does the fact that something happens to involve a computer automatically make it new and non obvious?
Just to be utterly clear here, hilighting mandatory fields on a paper form has been used for longer then computers exist.
Could you explain how Microsoft managed to get a patent on using xor to make a cursor blink? (this is an obvious method for anyone who ever did a little bit of assembler coding at the very least)
Can you explain the famous Amazon one click patent and it not being dismissed inmediately because of being utterly obvious? (this seems to be the most valid one of the 3 mentioned in this post actually)
Can you point us at any software patents that actually comply with the requirement of providing enough information so that a practiser in the field the patent applies to can reimplement it based on the text of the patent?
Thanks for your corrections but it still makes no sense at all.
You can selectively enforce a patent that has been granted, publication of patent applications does not vhange anything there because granted patents have to be published anyway, and can indeed be enforced selectively or not at all without becomming invalid.
Take a little peek maybe at how a company like IBM has been using and is still using patents, they do exactly what you say is impossible now, and do it with patents that have been granted and as such are public.
WHat you cannot do is try to keep a patent in limbo and as such unpublished for an extended period of time, but the sheer amount of patents out there, and the fact that in many cases people explicitly do not look for patents (knowingly infringing one is way more expensive then unknowingly doing so) makes that publication helps little if at all.
A vigorously enforced patent may get enough publicity to no longer be able to claim you did not know about it (some of the mp3 related patents, the now expired ones relating to the gif format etc for example)
> That is why BSD license is not as good as the GPL.
Guess it never occured to you that this is actually desired by those using the BSD license?
If forcing open source on anyone who derives from your work is your purpose then the GPL is better (this is quite legitimate, but not what everyone wants)
If making available reference implementations of a standard for anyone to use regardless of OSS is your purpose, then the BSD license is definitely better. (note that even people in the FSF suggest this)
Hmm, have been out of the loop with regards to modern phones for a bit, but my now ancient Ericsson T39m comes with a ringtone editor built in, pc software for editing ringtones, and PalmOS software for editing ringtones.. did things change that dramatically in the last few years?
> My wife's old nokia phone didn't have perfect reception here in Seoul, but it was usually alright.
Hehe, well, Nokia usually sucks when it comes to decent reception.. but hrm, many people have them or have something rather comparable.
> CDMA works perfectly fine here though, I can use my phone on a moving subway (and my phone is an old Samsung). I know there's a much smaller area to cover than the States, but it seems that even in major cities in the States, people have some problems with coverage. Which seems to be pretty unreasonable in this day and age.
Exactly.
> So instead, they just put on board members that will do the least damage from their perspective.
Thats obviously what they are doing. It is also exactly why many consider the current US government to be a fascist government, they comply very well with the original definition of it (note that this is not the same as a nazi government, which they are definitely not, also note that I am in no way comparing Bush to Hitler here, I am pointing at the original definition of fascism, which is basicly putting industry in charge of government. It is not going to be good for the people at all however, such governments without exception end up turning a substantial part of their population into slaves effectively)
> Protecting privacy reduces the power of government. Why does anyone expect any government to protect their privacy?
Maybe because they are there to serve the people and not the other way around?
Maybe because it is in the spirit of the constitution which they promissed to serve and protect?
> And your bashing of Bush may have gotten you mod points, but the Democrats are no slouches in the corporate malfeasance department. Terry McAuliff was up to his neck in Global Crossing, and the first person Ken DeLay called when the shit hit the fan at Enron was the Clinton administration's Treasury Secretary.
Sure, but republicans are supposedly advocating a lean and mean government with limited power, I have yet to hear democrats claiming the same.
Besides, 2 wrongs don't make a right. That the clinton admninistration messed up is no excuse whatsoever for Bush to mess up.
Hmm, do places like the Egyptian desert, jungle of northern Thailand or the mountains between France and Spain count as 'middle of nowhere' ? I found I could make and receive calls in those places without trouble whatsoever.
If coverage would be bad in areas with a relatively large human population that would be really bad, and I am sure that is no longer the case in the USA either, but honestly, when I read the stories here, or listen to my American friends who have cellphones, it looks like reading about how things were overhere more then a decade ago with regards to coverage.
I am currently sitting inside at my girlfriends house (in Berlin, Germany), and I have full 5 bar reception and perfect sound quality, and it may drop to 3 bars, but there is no place within an hour travel from here where inside coverage wont be good enough for reliably using a cellphone (and that includes all subway stations (which have repeaters here) and trains and office buildings and so on that I have been to)
The nearest place that I found without superb inside coverage is some 2 hours away from here by high speed train, and there you will find that the type of phone makes a big difference. My old Ericsson (T39m) with its external antenna still has a usable signal there, many cheap phones with internal antenna will no longer work reliably there when inside.
Ah yeah, having 4 or more networks to chose from (I can roam on all networks here) helps a bit there of course but in my 'home' area where I cannot use roaming, coverage is even better then overhere.
SA can query blacklists (and does in its default config) but, it uses those as an indication. If maps or xbl blocks an address, but all other indicators are saying it is not spam, SA will not mark the mail as spam normally. Of course you can configure that differently if you feel like it.
At any rate, rbls should only be used as one of the possible indicators for spam, none of the rbls is perfect or even uptodate, and reöying on them as the main or even only indicator for spam is just a very good way to block legitimate mail, while the effect on stopping spam is not even close to 70% (at least on the mail servers that I run, which serve a few hundred users each)
XBL seems to be one of the more usefull ones among the rbls, rbls aiming purely at dynamic/home IPs seem to be utterly useless in practise.