I'm sure that the schools wouldn't turn down some helpful (& free gratis) advice from a certified IT professional. Do you think the head and governors (or whatever education boards you have in the US) have, in the main, the first clue about setting up a school wide network for 1000 users?
If you do (have a clue), perhaps you could help out.
Yeah, I know often suppliers are centrally mandated and why would the school trust you - if you're a parent and you work for a well known (locally or nationally) company I think your chances are high.
Another thing (while I'm ranting): Schools have a lot of money to spend on equipment, perhaps if they were allowed to spend a bit more on consultation then they wouldn't waste all the equipment money!?
Well... it had been sick for so long, who knows if it will ever fully recover! And the other obvious retort that 'emacs is best'.
I use nano BTW, except when I have to edit sudoers or something when I spend half an hour hitting <esc> then a RandomKey and getting nothing done. Then I try to work out how to use nano instead. It's usually something simple, like using a different distro:0)>
How about setting the image size in your CSS according to relative font sizes (em for example)?
That way when people change the font size the images scale. Of course you have to be creative to have a nice looking site no matter how this affects the flow of text around the images.
Patent monopolies are only as broad as the area defined by the claims. I only looked at the eldest patent via Outlaw (http://www.out-law.com/php/page.php?page_id=10873 19985&area=news) to the USPTO (http://patft.uspto.gov/netacgi/nph-Parser?Sect1=P TO1&Sect2=HITOFF&d=PALL&p=1&u=/netahtml/srchnum.ht m&r=1&f=G&l=50&s1=5,694,546.WKU.&OS=PN/5,694,546&R S=PN/5,694,546).
The claim 1 (which is usually the broadest) specifies automatic download and uncompressing of files using a software component that can be embedded in an "vendor-provided containing information product". The embedded component also does all the dialup stuff and uses manifests to decide the file downloads.
The examiner appears to have been quite thorough, citing mirror-ing scripts for example.
So, it sounds alot like apt but as an embeddable component. The restrictions to "information file object[s]" is interesting as this appears to rule out executables. Also, if you use a pre-existing net connection (including a pre-dialled connection) it seems you are outside the claims. There's enough in the main claim (such as the requirement that it uses a "a communications network lacking a network file system" (there must be at least one NFS attached to the www, right?!)) means it's a pretty harmless claim.
Point at issue is that you can still be sued and it will probably be cheaper to pay for a license than to challenge in court!!
You mean just like CDs did...? Or perhaps you mean like nurofen (tradename for ibuprofen, granted it's more widespread since the patent lapsed, but it didn't die). Maybe, you mean that it will fade away like ring-pulls...
Just because something is protected by a patent doesn't mean that it can't be licensed reasonably. Rewarding good, genuinely innovative, ideas is OK in my book.
Of course, this is quite clever as it uses hardware as well as software and so can more easily be patented in places that restrict software patents (which is still true in Europe, whatever the press says).
Ming allows PHP to dynamically create shockwave flash movies (swf) either from randomised data or from "real" data. This is just something quite new being used in an interesting way.../.-ers get off [not literally I don't think] on this kind of thing.
You lose geek points for that post I think.
The ming webpages are worth a look. How about a site-meter that shows the number of visitors on site at present as individual beating hearts in a flash movie... cool!?
If you can do it in figures for PHP you can now make pretty movies abstracted from those figures.
But the impact of post-modernist thought on the art world has made it such that it's no longer important what the initial thought was... now the viewer decides, with post-modernism there is no objective concept. So, to spell it out... it doesn't matter what the artists concept is so long as the viewer perceives that there is one!
Of course there's a case for us having entered post-post-modernism but it's not clear on the complexion of that value system yet, IMHO.
PS: I'm an objectivist personally, I believe in objective truth, lot's of people don't appear to though.
Ooops, hit enter by mistake... anyway, I noticed that too, so I thought it meant the whole chicken-y bit was chicken breast. Having sampled the "new McNuggett" I reckon that it's still partially reconstituted / soya protein ; they didn't after all say that it was _only_ made with chicken breast pieces!
The noted prior art includes this http://arxiv.org/abs/cs/0008019 - which is an analysis of Bayesian approaches to spam filtering. It may well have been cited by the examiner but even so...
It's a bit much calling the patent examiner "inept" when you only skimmed the document. It's clear from looking at the claims and citations that the claims were narrowed considerably from merely a "bayesian spam filter", that everyone thinks is being claimed, to the current system. The claims are a compound of several spam filtering techniques along with use of a narrowly defined paragraph based hashing mechanism.
I've not looked into the techniques used but it doesn't seem like anyone is using this system.
No, I don't approve of software patents either. But this is far from a bad example.
hsoft $> "I don't see what's the point of hashing words"
Applying hash algorithms doesn't necessarily mean md5 or sha, its just a mathematical operation; usually we mean (pseudo) one-way hash but again, there are others. So it could be taking the first 4 letters of every word... see http://en.wikipedia.org/wiki/Hash_function for more.
Of course emails about shitake mushrooms with a lightly tossed salad might not make the grade:0)>
When you say "I'm utilizing a, b, c and d for doing X", it doesn't mean you actually use all of them.
It does if it's a patent document, they are legal claims for property. They have to be precise. For preferable features (ie optionally additional) the form is that subsequent "dependent" claims are used.
So, in this instance the parent post is correct. This is quite narrow.
PS: The parent post needs to use "Plain Old Text" option as well, and don't forget to preview!!!
This is almost what happens already... their is an opposition period (definitely in UK and Europe and I think in US too) in which you can file comments.
In UK these are observations (Section 21 of Patents Act) and don't require the observer to get involved. That is, you can write to the UKPO and say "this patent application is totally bogus, look at this RFC from 1932" and the examiner will take that into account. The applicant is notified and the document (I believe) if it's available will be placed on file.
Implementing a web-based version would make it difficult (I imagine) to extract proper comments from the background noise. Requiring postage costs cuts alot of the noise.
Of course who sits reading hundreds of patents a day just to check up on the USPTO? (other than businesses of course!).
Whilst the InfoWorld article quotes '"To me this looks like a pretty broad patent," said Rob Tosti' it doesn't look as broad as the headlines suggest.
The key feature is in claim 1...
"paragraph hashing by hashing a plurality of paragraphs and utilizing a database of hashes of paragraphs, wherein the paragraph hashing excludes at least one of a first paragraph and a last paragraph of content of the electronic mail messages wherein a plurality of hashes each has a level associated therewith, and the hashes having a higher level associated therewith are applied to the electronic mail messages prior to the hashes having a lower level associated therewith"
That's quite a tight restriction. If you're hashing the first and last paragraphs, for example, then you're in the clear! Of course this wouldn't stop them chasing you with a law-suit it would just mean you could be acquitted if you could afford to go the distance - [sarcasm] capitalism, I'm loving it!! [/sarcasm]
Also, I note that in the http://www.paulgraham.com/spam.html article hashing is only used in terms of words [yeah tokens really, but who's being pedantic?]. Here the restriciton of the claims is to hashing paragraphs. If you're hashing words you're OK (previous disclaimer applies!).
That's not to say that I think they deserve a patent. Just that the knee-jerk - "this is hugely broad" - isn't really justified IMHO.
I do object. I like some of their shows and approve of their ideology to improve the level of broadcasting and to increase the range of viewing available.
So much of it is trash though. And they advertise, just it's monopolistic... they only advertise BBC products (mags, books, videos,...).
Other things I object to: swapping Parkinson for Premiership football. They haven't increased the range of viewing, only taken a hugely overpriced show (association football [I live in Wales, where "football" weirdly means "rugby football"]) and charged it to the license payer instead of commerce. This is very bad, not least because it props up the price at public expense when fully commercial systems would bring down the price.
Oh, and I don't like paying for digital TV when I can't afford to buy a STB (#60 pounds-ish).
50 for a night out?? That's basically my disposable income for an entire month and 7.50 of that currently goes on my dial-up connection! Yes I am working (not at this precise second though:0)>
I thought that was what http://www.archive.org/web/freecache.php is for... you simply put a link as http://freecache.org/http://link.to/mysite.jpg ; freecache then retrieves a copy of the linked material and serves it from their site.
Apple don't have to patent it.. they can use defensive publication [*] (which will be alot cheaper than patenting). There are several established methods for doing this.. they could apply for a patent and not go beyond the publication stage of the patent or they could use a repository such as www.priorartdatabase.com.
Of course there are other options, patents with free (gratuit) licenses for example.
Then if anyone tries to haul Apple into court they can either point at the publication of the patent doc or some other date-verifiable publication to prove they have prior art.
The examiners job is not to produce 100% valid patents (which most companies probably don't want; democracy-s-mockery) but to produce patents that are _likely_ to be valid : the difference is a few hundred man-years of searching and high resources drain on the economy versus (at most) a couple of man-days of searching.
-- * defensive publication means producing the matter of an invention in such a way that the date of it's disclosure can be easily verified and preferably the disclosure can be readily accessed by patent searchers.
Quantum effects don't appear to happen to items above the Plank scale (ie only affects small things - sub-atomic).
What is worse is that items such as photons and electrons are considered to be point items with no appreciable size, they are irreducible (not made of smaller parts) so bits can't fall off.
How we see things at the macroscopic scales of everyday life is hard to relate to the quantum scale... items are both waves _and_ particles at the same time (depending on the interpretation you choose to follow, but I'll gloss over that).
To put it another way, items are particles that have a probability distribution in space and time, they might be here, they might more likely be there. The diffraction pattern (the distribution of VWs on the screen) is caused by the probability of any particle appearing in a particular place being similar for all the particles... their are points of likelyhood for particles to impact on. All the particles are identical in form. It is not to do with differences in the particles make-up.
This, incidentally, introduces an indeterminancy into subatomic interactions ("will this particular electron form that part of the pattern or that part?").
And so a possible description (from the article) then appears to be that each electron is a multitude of electrons in parallel universes forming all parts of the pattern at once but that only one of those multiverse electrons is visible to us. This appears to be philosophy however as there is no clear test that can refute it.
Wibble wibble bing... back to work.
[Yes I did quantum mechanics and quantum field theory at Uni, I wasn't very good at it but I enjoyed it. My 'field' (no pun intended) of preference is particle physics.]
MG's is presumably machine-guns... hmm, yeah, on actually _reading_ (rather than my usual scan-read thing) the article more closely it does say multibarrels for the million rounds. Can't this be achieved with any gun as long as you have enough barrels??!
What a cheat, specially when the headline is "A GUN THAT FIRES A MILLION ROUNDS A MINUTE"... what they mean is a set of guns. I suppose 60,000 rounds a minute didn't sound impressive enough. Other flaws include the rounds being up to 60mm (6cm) diameter - though I expect they aren't fired at such high rates.
Also, yeah, it's not electric that's used to power the bullets flight but I didn't bother to look up the energy provided by propellant explosives.
When you shoot this gun (I did look at the article and there aren't any details except that it's electric) more than 15000 bullets a second are leaving the muzzle. If each bullet is 1cm in length that's at least 150m of bullet and assuming a recycle time equal to 10 times the length of the bullet [*] let's say 3000m. That's a firing velocity of Mach 9.
Also, a one million strong line-up of 1cm bullets adds up to 10km of metal being fired each minute! Alternatively if each bullet is 1cm^3 of metal that's a m^3 of metal which is likely to weigh in excess of 7 metric tons (using Iron, 7380 kg/m^3 as a guidline).
So each 60seconds we accelerate 7+ tons of bullet metal to Mach 9... using KE (kinetic energy) formula we give it 30 thousand-million Joules of energy @ 500 million Watts (about the output of 5 large electric plants).... They're going to need big batteries!
[*> that is the bullet has moved ten times it's length before the next bullet sets off]
PS: I'm sure someone will find a mistake in these calculations and that someone else with more gun knowledge will correct some horrible assumption, but hey.
Slashdot Mirror
I'm sure that the schools wouldn't turn down some helpful (& free gratis) advice from a certified IT professional. Do you think the head and governors (or whatever education boards you have in the US) have, in the main, the first clue about setting up a school wide network for 1000 users?
If you do (have a clue), perhaps you could help out.
Yeah, I know often suppliers are centrally mandated and why would the school trust you - if you're a parent and you work for a well known (locally or nationally) company I think your chances are high.
Another thing (while I'm ranting): Schools have a lot of money to spend on equipment, perhaps if they were allowed to spend a bit more on consultation then they wouldn't waste all the equipment money!?
"Vi is better"
... it had been sick for so long, who knows if it will ever fully recover! And the other obvious retort that 'emacs is best'.
:0)>
Well
I use nano BTW, except when I have to edit sudoers or something when I spend half an hour hitting <esc> then a RandomKey and getting nothing done. Then I try to work out how to use nano instead. It's usually something simple, like using a different distro
I'm kidding!!!
How about setting the image size in your CSS according to relative font sizes (em for example)?
That way when people change the font size the images scale. Of course you have to be creative to have a nice looking site no matter how this affects the flow of text around the images.
Read The Flippin' Claims ...
3 19985&area=news) to the USPTO (http://patft.uspto.gov/netacgi/nph-Parser?Sect1=P TO1&Sect2=HITOFF&d=PALL&p=1&u=/netahtml/srchnum.ht m&r=1&f=G&l=50&s1=5,694,546.WKU.&OS=PN/5,694,546&R S=PN/5,694,546).
Patent monopolies are only as broad as the area defined by the claims. I only looked at the eldest patent via Outlaw (http://www.out-law.com/php/page.php?page_id=1087
The claim 1 (which is usually the broadest) specifies automatic download and uncompressing of files using a software component that can be embedded in an "vendor-provided containing information product". The embedded component also does all the dialup stuff and uses manifests to decide the file downloads.
The examiner appears to have been quite thorough, citing mirror-ing scripts for example.
So, it sounds alot like apt but as an embeddable component. The restrictions to "information file object[s]" is interesting as this appears to rule out executables. Also, if you use a pre-existing net connection (including a pre-dialled connection) it seems you are outside the claims. There's enough in the main claim (such as the requirement that it uses a "a communications network lacking a network file system" (there must be at least one NFS attached to the www, right?!)) means it's a pretty harmless claim.
Point at issue is that you can still be sued and it will probably be cheaper to pay for a license than to challenge in court!!
When you can taste an oncoming truck you know you're in trouble!!
You mean just like CDs did ...? Or perhaps you mean like nurofen (tradename for ibuprofen, granted it's more widespread since the patent lapsed, but it didn't die). Maybe, you mean that it will fade away like ring-pulls ...
Just because something is protected by a patent doesn't mean that it can't be licensed reasonably. Rewarding good, genuinely innovative, ideas is OK in my book.
Of course, this is quite clever as it uses hardware as well as software and so can more easily be patented in places that restrict software patents (which is still true in Europe, whatever the press says).
pbhj
How can you see my hair from all the way over there ...? Must be this new inter-m-nat thingy??!
Ming allows PHP to dynamically create shockwave flash movies (swf) either from randomised data or from "real" data. This is just something quite new being used in an interesting way ... /.-ers get off [not literally I don't think] on this kind of thing.
... cool!?
You lose geek points for that post I think.
The ming webpages are worth a look. How about a site-meter that shows the number of visitors on site at present as individual beating hearts in a flash movie
If you can do it in figures for PHP you can now make pretty movies abstracted from those figures.
HTH
pbhj
But the impact of post-modernist thought on the art world has made it such that it's no longer important what the initial thought was ... now the viewer decides, with post-modernism there is no objective concept. So, to spell it out ... it doesn't matter what the artists concept is so long as the viewer perceives that there is one!
Of course there's a case for us having entered post-post-modernism but it's not clear on the complexion of that value system yet, IMHO.
PS: I'm an objectivist personally, I believe in objective truth, lot's of people don't appear to though.
Ooops, hit enter by mistake ... anyway, I noticed that too, so I thought it meant the whole chicken-y bit was chicken breast. Having sampled the "new McNuggett" I reckon that it's still partially reconstituted / soya protein ; they didn't after all say that it was _only_ made with chicken breast pieces!
:0)>
Marketeers are evil!
The noted prior art includes this http://arxiv.org/abs/cs/0008019 - which is an analysis of Bayesian approaches to spam filtering. It may well have been cited by the examiner but even so ...
I'd have thought that you might miss some junk (possibly designed to fool filters)?
OK.
1 6171 ]
It's a bit much calling the patent examiner "inept" when you only skimmed the document. It's clear from looking at the claims and citations that the claims were narrowed considerably from merely a "bayesian spam filter", that everyone thinks is being claimed, to the current system. The claims are a compound of several spam filtering techniques along with use of a narrowly defined paragraph based hashing mechanism.
I've not looked into the techniques used but it doesn't seem like anyone is using this system.
No, I don't approve of software patents either. But this is far from a bad example.
[Cross reference http://slashdot.org/comments.pl?sid=109666&cid=93
hsoft $> "I don't see what's the point of hashing words"
... see http://en.wikipedia.org/wiki/Hash_function for more.
:0)>
Applying hash algorithms doesn't necessarily mean md5 or sha, its just a mathematical operation; usually we mean (pseudo) one-way hash but again, there are others. So it could be taking the first 4 letters of every word
Of course emails about shitake mushrooms with a lightly tossed salad might not make the grade
It does if it's a patent document, they are legal claims for property. They have to be precise. For preferable features (ie optionally additional) the form is that subsequent "dependent" claims are used.
So, in this instance the parent post is correct. This is quite narrow.
PS: The parent post needs to use "Plain Old Text" option as well, and don't forget to preview!!!
This is almost what happens already ... their is an opposition period (definitely in UK and Europe and I think in US too) in which you can file comments.
In UK these are observations (Section 21 of Patents Act) and don't require the observer to get involved. That is, you can write to the UKPO and say "this patent application is totally bogus, look at this RFC from 1932" and the examiner will take that into account. The applicant is notified and the document (I believe) if it's available will be placed on file.
Implementing a web-based version would make it difficult (I imagine) to extract proper comments from the background noise. Requiring postage costs cuts alot of the noise.
Of course who sits reading hundreds of patents a day just to check up on the USPTO? (other than businesses of course!).
Whilst the InfoWorld article quotes '"To me this looks like a pretty broad patent," said Rob Tosti' it doesn't look as broad as the headlines suggest.
...
The key feature is in claim 1
"paragraph hashing by hashing a plurality of paragraphs and utilizing a database of hashes of paragraphs, wherein the paragraph hashing excludes at least one of a first paragraph and a last paragraph of content of the electronic mail messages wherein a plurality of hashes each has a level associated therewith, and the hashes having a higher level associated therewith are applied to the electronic mail messages prior to the hashes having a lower level associated therewith"
That's quite a tight restriction. If you're hashing the first and last paragraphs, for example, then you're in the clear! Of course this wouldn't stop them chasing you with a law-suit it would just mean you could be acquitted if you could afford to go the distance - [sarcasm] capitalism, I'm loving it!! [/sarcasm]
Also, I note that in the http://www.paulgraham.com/spam.html article hashing is only used in terms of words [yeah tokens really, but who's being pedantic?]. Here the restriciton of the claims is to hashing paragraphs. If you're hashing words you're OK (previous disclaimer applies!).
That's not to say that I think they deserve a patent. Just that the knee-jerk - "this is hugely broad" - isn't really justified IMHO.
pbhj
Hmm,
... they only advertise BBC products (mags, books, videos, ...).
:0)>
I do object. I like some of their shows and approve of their ideology to improve the level of broadcasting and to increase the range of viewing available.
So much of it is trash though. And they advertise, just it's monopolistic
Other things I object to: swapping Parkinson for Premiership football. They haven't increased the range of viewing, only taken a hugely overpriced show (association football [I live in Wales, where "football" weirdly means "rugby football"]) and charged it to the license payer instead of commerce. This is very bad, not least because it props up the price at public expense when fully commercial systems would bring down the price.
Oh, and I don't like paying for digital TV when I can't afford to buy a STB (#60 pounds-ish).
50 for a night out?? That's basically my disposable income for an entire month and 7.50 of that currently goes on my dial-up connection! Yes I am working (not at this precise second though
pbhj
I can't believe you actually hacked wikipedia just to give some weight to your argument ... they must pay you alot to work at M$. ;0)>
Oh.
You could pad them!?
No, I don't think you should either.
I thought that was what http://www.archive.org/web/freecache.php is for ... you simply put a link as http://freecache.org/http://link.to/mysite.jpg ; freecache then retrieves a copy of the linked material and serves it from their site.
Brilliant!
Apple don't have to patent it .. they can use defensive publication [*] (which will be alot cheaper than patenting). There are several established methods for doing this .. they could apply for a patent and not go beyond the publication stage of the patent or they could use a repository such as www.priorartdatabase.com .
Of course there are other options, patents with free (gratuit) licenses for example.
Then if anyone tries to haul Apple into court they can either point at the publication of the patent doc or some other date-verifiable publication to prove they have prior art.
The examiners job is not to produce 100% valid patents (which most companies probably don't want; democracy-s-mockery) but to produce patents that are _likely_ to be valid : the difference is a few hundred man-years of searching and high resources drain on the economy versus (at most) a couple of man-days of searching.
--
* defensive publication means producing the matter of an invention in such a way that the date of it's disclosure can be easily verified and preferably the disclosure can be readily accessed by patent searchers.
The analogy is fundamentally flawed:
... items are both waves _and_ particles at the same time (depending on the interpretation you choose to follow, but I'll gloss over that).
... their are points of likelyhood for particles to impact on. All the particles are identical in form. It is not to do with differences in the particles make-up.
... back to work.
Quantum effects don't appear to happen to items above the Plank scale (ie only affects small things - sub-atomic).
What is worse is that items such as photons and electrons are considered to be point items with no appreciable size, they are irreducible (not made of smaller parts) so bits can't fall off.
How we see things at the macroscopic scales of everyday life is hard to relate to the quantum scale
To put it another way, items are particles that have a probability distribution in space and time, they might be here, they might more likely be there. The diffraction pattern (the distribution of VWs on the screen) is caused by the probability of any particle appearing in a particular place being similar for all the particles
This, incidentally, introduces an indeterminancy into subatomic interactions ("will this particular electron form that part of the pattern or that part?").
And so a possible description (from the article) then appears to be that each electron is a multitude of electrons in parallel universes forming all parts of the pattern at once but that only one of those multiverse electrons is visible to us. This appears to be philosophy however as there is no clear test that can refute it.
Wibble wibble bing
[Yes I did quantum mechanics and quantum field theory at Uni, I wasn't very good at it but I enjoyed it. My 'field' (no pun intended) of preference is particle physics.]
MG's is presumably machine-guns ... hmm, yeah, on actually _reading_ (rather than my usual scan-read thing) the article more closely it does say multibarrels for the million rounds. Can't this be achieved with any gun as long as you have enough barrels??!
... what they mean is a set of guns. I suppose 60,000 rounds a minute didn't sound impressive enough. Other flaws include the rounds being up to 60mm (6cm) diameter - though I expect they aren't fired at such high rates.
What a cheat, specially when the headline is "A GUN THAT FIRES A MILLION ROUNDS A MINUTE"
Also, yeah, it's not electric that's used to power the bullets flight but I didn't bother to look up the energy provided by propellant explosives.
When you shoot this gun (I did look at the article and there aren't any details except that it's electric) more than 15000 bullets a second are leaving the muzzle. If each bullet is 1cm in length that's at least 150m of bullet and assuming a recycle time equal to 10 times the length of the bullet [*] let's say 3000m. That's a firing velocity of Mach 9.
... using KE (kinetic energy) formula we give it 30 thousand-million Joules of energy @ 500 million Watts (about the output of 5 large electric plants). ... They're going to need big batteries!
Also, a one million strong line-up of 1cm bullets adds up to 10km of metal being fired each minute! Alternatively if each bullet is 1cm^3 of metal that's a m^3 of metal which is likely to weigh in excess of 7 metric tons (using Iron, 7380 kg/m^3 as a guidline).
So each 60seconds we accelerate 7+ tons of bullet metal to Mach 9
[*> that is the bullet has moved ten times it's length before the next bullet sets off]
PS: I'm sure someone will find a mistake in these calculations and that someone else with more gun knowledge will correct some horrible assumption, but hey.