Funny... I have very limited exposure to CSS and I don't have any issues with what you are talking about. Dynamic positioning (Relative, I think is what you are saying) makes sense if you just think about math for a bit, and absolute positioning is good for a lot of things (headers, iconboxes, etc.) and I think you miss the whole point of cascading.
I've built about 3 full-scale apps with web-interfaces to them as well, and they all have used CSS and all work just fine. PEBCAK...
Is there an opposite of the disorder? I absorb massive quantities of caffeine without getting hyper, and I can code for days on a single project with just the occasional break for more caffeine or food.
I'm sort of the same way. When it comes to a few things my attention span is great. Elsewhere, I can't focus very well. Programming, driving (not commuting, driving), car racing games (see previous) I can sit there for days and be completely absorbed.
When I was in school, reading a book that sucked but was required reading took an act of God.
As a matter of principle, whenever I install software I go out of my way to avoid having to click "accept". Often this means hand-unpacking the contents of the CD, or using a no-CD patch. Presumably (assuming I haven't violated the DMCA in the process), I am then bound by copyright law with respect to the software. But then the vast majority of software I use is free, so I'm not affected very much by piles of shrink-wrap licenses.
This is why they usually have such agreements on the jewel case, with conspicious text saying "You agree if you open this case." Easier to prove that way, because you either a) violated the license or b) violated copyright.
Not all EULAs are bad, for instance at my job we have a EULA that must be agreed to otherwise we are in violation of certain academic regulations. Our EULA contains refund information, which must be presented and agreed to. If you don't agree, we can't do business with you. That's our EULA though, aside from the copyright jargon which is so vanilla you could eat a brownie with it.
Our EULA can be easily proven, because it's stored on the server side and we track IP and login information.
I hate the associations that certain things are always evil. EULA's can be good, and bad. Just because most of them are bad, doesn't make them all bad. It's like people, I know a lot of asshole (white|black|asian) people but it doesn't mean they're all like that... just the majority;)
Think of a video rental model, except that you do not have to return the media!
You mean Divx, and Circuit City got burned hard on that. My point was completely true, since I explicitely stated "unless I can re-use the media it came on" meaning that the media is mine and mine alone. I physically own it. I do not own the data and content on that media.
As far as I know, this has not yet been upheld by any court as a valid means of establishing a contract. (And I hope that we can get "shrink wrap" contracts invalidated) If you have a counter-example, I'd be interested to hear.
Shrink-wrap licenses, no. There is major precedent for unsigned contracts being upheld. If you say you will do something, and you don't do it you violated contract law. Just as long as there is evidence enough to support the claim.
If I ever have software that will be supplied that needs a restrictive license, I will collect signatures. Otherwise, I think copyright law is sufficient for my needs. Many others don't feel that way.
Music is governed by copyright law, not contract law. Copyright law gives you the right to do many things that the media conglomorates don't want you to know about (like making backups, i.e. "fair use"). There is no contract involved when purchasing a CD. What about the CDs that now come with EULAs? I don't buy new CDs, because of my "don't buy it, don't bitch" stance. I buy used CDs so I have no idea. I do know there are a few that have EULAs. IF a CD has a EULA, it also is bound by copyright and contract law.
The attempt to subsume copyright law into contract law is at the heart of all these DRM attempts. However, if you do not agree to the contract, have no opportunity to negotiate the contract, or both parties do not sign the contract, it is not valid.
Well, if both parties do not agree to the contract it is not valid. If, in bold letters, I put "By opening this package you agree to the following license." it would be reasonable to assume that you agreed. Not all contracts require a signature.
DRM is completely engrossed in using contract law because copyright is inadequate (to what they are trying to do.)
Contracts which violate other rights are simply illegal contracts. (You cannot, for instance, sign a contract allowing someone to kill you -- it's still murder and they will still go to jail)
That is incorrect, contracts can violate other rights (specifically copyright, limiting those rights) and are perfectly legal. The difference is looking at if it's a right-reduction/limitation. If you sign a contract and commit a felony in the bounds of that contract, you still commited a felony. If you sign a contract removing all of your rights to something, there is no felony. This is what drives employment contracts that revoke rights to work on external projects, and if you do you must disclose all information to the company/organization.
It's dangerous to confuse contract law, which is civil and criminal law (the murder example.)
If you agree to a contract with me, and violate that I sue you. If you violate a state law (criminal law) than the state will take you to court, and I still have the right to do so. A great example of this is the O.J. Simpson case. He was acquitted in the criminal case, but was found guilty in the civil wrongful death case.
I'm not a lawyer, I just spend too much time studying it...
I still haven't bought a DVD player because they can't arbitrarily fast-forward and rewind. I'm not bitching, but it seems like if this were really a free market economy, somebody would come along to sell me an unrestricted DVD player. And I mean "unrestricted" as an advertised feature, not some chip I have to solder in.
I have a Sony. When I put a DVD in I hit "Menu" then "Play" immediately. It then starts playing with no restrictions. It's up to the people who make the discs to enable or disable certain features in the playback.
It has nothing to do with the DVD player (and a lot of DVD players you can actually control via software, just google for it) but the disc.
then I because a citizen and the government controlled me.
Point.
Now I'm a consumer, and all my rights are under control.
Excercise the right not to buy. Didn't that thought ever occur to anybody?
Making rip-proof media is fine, if they can figure out how to do it. You are agreeing to a contract in regards to the content on whatever media you are purchasing. Self-destructing is destroying my property, unless I can re-use the media it came on.
If I don't like what a company is doing, I will not buy their goods. I still don't buy Colgate products because of them bullying ajax.org.
Stop pretending like you don't have rights. You have rights, you chose not to use them. You are a consumer, and you will buy everything they tell you to buy. Then you bitch when they take you all for suckers.
If you don't like it, don't buy it. If you do buy it, don't bitch.
Don't get me wroing, I don't love AIX by any stretch of the imagination;-) But this is starting to seem like the technology equivalent of Days of our Lives or something!
AIX is great, if it's used for what it's designed for... but that's not why I'm responding.
I see this as the equivalent to Celebrity Deathmatch, except those episodes where it is a stale, crusty has-been going against a massive gorilla.
>>Do I _REALLY_ want to pull in libpng and libSDL just to do this?
>Yes.
Uhm, to display a PNG to the screen. I can understand pulling libpng, but not libSDL.
It sounds like you either have a delusionally high estimation of your own abilities relative to those of your library-writing peers ("Of course the code I write will be faster, more correct, and less bloaty than the specialised code that those people who really understand the problem and spend time solving it well!") or you place very little value on your time. Oh well. Have fun writing png-frobbing code for the rest of your life.
Most open source code is shit. It's not because there aren't great programmers, it's because there are a shitload of bad programmers. That whole 80/20 thing. Unless you are familiar with a library, I don't blame anybody for not using it for small projects.
If I was writing a png viewer, I'd do it with libpng and X. X isn't that hard, it's a pain in the ass to work with for large programs, but it's not that hard.
Big Deal. Hotel rooms should basically have a nice bed and nice bath. Personally, if I am going on vacation, I want to spend as much time OUT of the hotel room as possible but when I come back I want to get a good nights sleep and get clean the next morning. This type of stuff is designed to attract the people who bought those old Acer computers just because they had a black case.
You are completely and totally wrong. There are tons of reasons why this is a good thing. Executive travelling, where they aren't on vacation but on a business trip. Second, visiting friends on vacation to a place where you know the area, and don't need to site see but just hang out with your friends.
This type of stuff is designed to attract people who have money, and want to stay in comfort.
Just because you don't understand something, doesn't mean it's bad/stupid/wrong. It just means you don't understand the need for it.
I guess you also don't understand why people like to own a BMW, or Mercedes.
North America and Europe were the only two continents industrialized enough to need that amount of oil.
You seem to be forgetting those little events called World War.
So, we showed them it was there (all they saw was a vast, useless desert), gave them the technology to extract it, and gave them the markets to sell their otherwise useless oil to.
As opposed to be conquered by Germany, and having it stolen and their people raped and murdered... I can see why they hate the US and Britain...
It's unlikely that more than 1-2% of US desktops are running a Free operating system.
It was unlikely in 1995. You are forgetting that there has been Lindows, huge marketing pushes, and tons of geeks doing installfests. I would say it's closer to 5%.
As for poor quality software, I suppose you haven't used BIND or Sendmail, eh? Even "better" software (Apache, Samba, OpenSSH, etc.) still has remote root holes not too uncommonly, and the Linux kernel has had hundreds of local root holes.
This is just a dumb argument, nobody calls their own child ugly.
What kind of language model did you use for the IRC analyser? It would probably possible to create a model resilient to bad spelling.
Just statistics based upon word relations, ignoring certain prepositions. I was testing an idea that I had to determine if you could find the difference between languages (like SVO/SOV syntaxes) through statistical analysis.
You can also check my JE, I post in there quite a bit.
I'm not sure it will work very well. In fact I doubt it will, information extraction works best for huge corpuses of text with some fairly simple regularity in their structure. It can still be natural language, but one might make the observation that A is usually before B, and bias the system for such a special case.
Just for an algorithm idea, if you scan a huge glob of text (web-pages, etc) and form patterns based upon word relevance and relation, and then categorize the webpage (manually or through automated tools... screenscarping google, etc.) and have a final result with loose attachments to category. This is just for building the statistics engine, however. I was working on an idea for this using IRC text, but considering most of the people on IRC type with the acumen of autistic lepars, it didn't turn out to well.
There was one project where they mined medical journal articles for which genes were found to affect which properties. I don't have a link to the paper right now but if you're interested I can look it up.
Ironic... after my NN work I went to work writing genetic sequence analysers.. Mostly finding patterns in the contigs to find A/T stops (much more challenging than it seems, especially when a 0% false-positive rate has to be there)
I might put up a website at some point. The project is both school work and commercial work so I'll get around doing it at some point. The interesting part is school work and the useful part (the application) is commercial.
We should talk more, if you have the time/desire I am pretty sure I'd love to bring you into nerdfarm. Just a general mish-mash of open source work, some of it is theoretical. We're currently building the website framework right now so there isn't anything to see on the page.
Perhaps you should just familiarize yourself a bit more with UNIX command line utilities; the ease and power with which skilled UNIX users can find and organize things is one of the major selling points of UNIX.
Uhm, typical "I'm a l33t l33n0x" response. Thanks for wasting your time on an attempt to try to make yourself seem more elite. If you had actually read what I was going for, you would understand that your system of symlinking and fgrep lines makes you look like a fucking retard. Like this:
Organizing data is hard, but chances are that whatever you want to do is already pretty easy with existing tools in UNIX/Linux; otherwise, people would have already added tools for addressing the problem long ago.
Because Haystack is doing something different than what is done, and Mac OSX is working towards the same goal. Linux just clones what the competition is doing, a few years later.
Thanks for playing, save your elitism for high school.
The fact that neither of these terms have shown up in the discussion thus far indicate how little business training most slashbots have. See, sometimes, there's value to an MBA. If nothing else, we're more conversent with lawyer-ese than most programmers.
I knew there were the terms, but couldn't remember what they were. Thanks;) I didn't realize you were an MBA... I could definitely use some of your help if you are interested.
(And in case you were wondering what all this means, basically, anyone who downloaded the software might have it free and clear. AOL, via Nullsoft, via authority of Justin, may have to abide by the original terms, GPL.)
That was what I was getting at. If the deal that AOL has over Nullsoft allows Nullsoft to act and release software independantly of the AOL label (such as Winamp, which afaik, is still Nullsoft Winamp, not AOL Winamp) than Justin could authorize the release under the GPL (excluding the mess with RSAREF.)
Bayesian filtering is another of those words that mean a lot of things;)
Isn't it just...
A widely used set of language models are the Hidden Markov Models. I'm planning to use them on an information extraction problem (populating database tables from free-text descriptions), and that's about the closest to the problem we're discussing that I've been. You could probably use them as a partial solution here as well, but I can't think of any really clever scheme at the moment.
Do you have a website or anything for this? It seems pretty cool, we should talk. If you need webspace or a server for it I can provide it.
And finally, sorry to be boring, but I'm not currently working on anything that would create something like the system we've discussed =)
I'm enjoying the conversation, I wish I had more time to dedicate to it but I am working on something... that isn't like what we're discussing at all, and is ultimately boring. As the saying goes, "Character is doing something that needs done with the interest in doing it fades away." At least we get to put in some cool magic in the backend.
What I would like to have would be something that could integrate all information related to some "project" into a single, powerful interface. Something that would allow me to have information in different formats (txt, html, pdf, ps, doc, spreadsheets, images and so on) searchable from the same locations. Also it would need to have a convenient way of "commenting" material written by others and linking these commments to the right place in the right document, without me having to place e.g. HTML links in the right locations. Then discussion material (mailing-lists, forums) should also be integrated in the same interface.
For me, this would be the ultimate window manager. When I'm on a computer, I'm generally working... or playing games but then I'm on Windows... so anyway. Having a truly integrated desktop would be awesome, but unfortunately too messy (especially with Linux) because there is no standard. If people were to actually come up with and adhere to a standard application model, something like this would be possible. Until that day, you'd have to re-invent so many wheels it wouldn't be feasible.
I guess my main point is that different media should not be artificially classified by its media type but rather by its contents (or more specifically which project it belongs to). My secondary point would then be that everything should be searchable, rather than found using a URL. More like the web and a relational database and less like a filesystem tree. Exactly - and that's what I would like. Even if I have to at first "train it" what things look like.
In regards to plain text/personal information -- have you thought about looking at Bayesian filtering for a solution to that? I haven't (yet) but the idea is festering in my brain.
is extended ext2 attributes (google for the kernel patch and userland utils) It allows you to give attributes to files, (like "category") I understand there's also gnome support.
Index the attributes and put limited statistical information checks on it (also access count, would be nice) and yeah... pretty much.
I might misunderstand you, but does this not mean that the "AI"-methods used by Haystack would not be AI?
Using AI methods does not constitute AI. I haven't looked at their design papers so I can't say for sure. I have yet to see a real AI implementation, so I would say they use AI-Methods, but not AI.
I also have some Neural Network / Statistical processing background so I tend to share this same view that the intelligence is in the designer not in the program. However, the term AI is often used to describe both neural and statistical methods.
Most people view AI as some ambigious concept that they can't understand, so they just slap the label AI on anything that involves automated decision making... *shrug*
Actually neural networks could also be perceived as statistical methods. I don't know about fuzzy logic but my intuition tends hint that they arent fundamentally very different either, just based on different mathematical theory.
Fuzzy Logic is massively statistical, just a bunch of true-false statistics.
To return to your relational filesystem idea, I would prefer having both relational search capabilities and google-like utilization of implicit information. I don't know about you, but I certainly am not very good at being consistent, so search and selection methods allowing some kind of fuzziness criteria would certainly be nice. The basic OS could still work with files, but my personal information should be organized in some more practical way, and should preferably be available all the time from diverse clients (home computer, work computer, PDA, cellphone...).
I'm consistent by putting them in directories... that's about as far as it goes. I try to organize it like $DEVROOT/(Commercial|Contract|OpenSource)/$Categor y That never really goes through... being able to store everything and pull it up using different methods would kick ass. Sometimes I write a perl script for Project "Oops, I forgot" that would be great in my current project. Digging through all of my source takes ages, but having something to figure out A) It's a perl script (easy), B) roughly what it's purpose is (hard) would be awesome.
Slashdot Mirror
Funny... I have very limited exposure to CSS and I don't have any issues with what you are talking about. Dynamic positioning (Relative, I think is what you are saying) makes sense if you just think about math for a bit, and absolute positioning is good for a lot of things (headers, iconboxes, etc.) and I think you miss the whole point of cascading.
I've built about 3 full-scale apps with web-interfaces to them as well, and they all have used CSS and all work just fine. PEBCAK...
I just want a phone that fits easily in the pocket of my jeans (has to coexist with my keys) and, pay attention now, gets good reception.
Good for you. This story isn't for you, it's for people who want phones that double as a PDA and messaging system.
Thank you for letting us know you don't want one though.
CSS is great for applying styles. It's layout mechanism sucks ass.
Uhm... you obviously don't have any experience with CSS. DIV layers provide the best positioning system possible within a webpage.
Just because you don't know how to do it, doesn't mean it sucks ass.
Is there an opposite of the disorder? I absorb massive quantities of caffeine without getting hyper, and I can code for days on a single project with just the occasional break for more caffeine or food.
I'm sort of the same way. When it comes to a few things my attention span is great. Elsewhere, I can't focus very well. Programming, driving (not commuting, driving), car racing games (see previous) I can sit there for days and be completely absorbed.
When I was in school, reading a book that sucked but was required reading took an act of God.
As a matter of principle, whenever I install software I go out of my way to avoid having to click "accept". Often this means hand-unpacking the contents of the CD, or using a no-CD patch. Presumably (assuming I haven't violated the DMCA in the process), I am then bound by copyright law with respect to the software. But then the vast majority of software I use is free, so I'm not affected very much by piles of shrink-wrap licenses.
;)
This is why they usually have such agreements on the jewel case, with conspicious text saying "You agree if you open this case." Easier to prove that way, because you either a) violated the license or b) violated copyright.
Not all EULAs are bad, for instance at my job we have a EULA that must be agreed to otherwise we are in violation of certain academic regulations. Our EULA contains refund information, which must be presented and agreed to. If you don't agree, we can't do business with you. That's our EULA though, aside from the copyright jargon which is so vanilla you could eat a brownie with it.
Our EULA can be easily proven, because it's stored on the server side and we track IP and login information.
I hate the associations that certain things are always evil. EULA's can be good, and bad. Just because most of them are bad, doesn't make them all bad. It's like people, I know a lot of asshole (white|black|asian) people but it doesn't mean they're all like that... just the majority
Think of a video rental model, except that you do not have to return the media!
You mean Divx, and Circuit City got burned hard on that. My point was completely true, since I explicitely stated "unless I can re-use the media it came on" meaning that the media is mine and mine alone. I physically own it. I do not own the data and content on that media.
As far as I know, this has not yet been upheld by any court as a valid means of establishing a contract. (And I hope that we can get "shrink wrap" contracts invalidated) If you have a counter-example, I'd be interested to hear.
Shrink-wrap licenses, no. There is major precedent for unsigned contracts being upheld. If you say you will do something, and you don't do it you violated contract law. Just as long as there is evidence enough to support the claim.
If I ever have software that will be supplied that needs a restrictive license, I will collect signatures. Otherwise, I think copyright law is sufficient for my needs. Many others don't feel that way.
Music is governed by copyright law, not contract law. Copyright law gives you the right to do many things that the media conglomorates don't want you to know about (like making backups, i.e. "fair use"). There is no contract involved when purchasing a CD.
What about the CDs that now come with EULAs? I don't buy new CDs, because of my "don't buy it, don't bitch" stance. I buy used CDs so I have no idea. I do know there are a few that have EULAs. IF a CD has a EULA, it also is bound by copyright and contract law.
The attempt to subsume copyright law into contract law is at the heart of all these DRM attempts. However, if you do not agree to the contract, have no opportunity to negotiate the contract, or both parties do not sign the contract, it is not valid.
Well, if both parties do not agree to the contract it is not valid. If, in bold letters, I put "By opening this package you agree to the following license." it would be reasonable to assume that you agreed. Not all contracts require a signature.
DRM is completely engrossed in using contract law because copyright is inadequate (to what they are trying to do.)
Contracts which violate other rights are simply illegal contracts. (You cannot, for instance, sign a contract allowing someone to kill you -- it's still murder and they will still go to jail)
That is incorrect, contracts can violate other rights (specifically copyright, limiting those rights) and are perfectly legal. The difference is looking at if it's a right-reduction/limitation. If you sign a contract and commit a felony in the bounds of that contract, you still commited a felony. If you sign a contract removing all of your rights to something, there is no felony. This is what drives employment contracts that revoke rights to work on external projects, and if you do you must disclose all information to the company/organization.
It's dangerous to confuse contract law, which is civil and criminal law (the murder example.)
If you agree to a contract with me, and violate that I sue you. If you violate a state law (criminal law) than the state will take you to court, and I still have the right to do so. A great example of this is the O.J. Simpson case. He was acquitted in the criminal case, but was found guilty in the civil wrongful death case.
I'm not a lawyer, I just spend too much time studying it...
I still haven't bought a DVD player because they can't arbitrarily fast-forward and rewind. I'm not bitching, but it seems like if this were really a free market economy, somebody would come along to sell me an unrestricted DVD player. And I mean "unrestricted" as an advertised feature, not some chip I have to solder in.
I have a Sony. When I put a DVD in I hit "Menu" then "Play" immediately. It then starts playing with no restrictions. It's up to the people who make the discs to enable or disable certain features in the playback.
It has nothing to do with the DVD player (and a lot of DVD players you can actually control via software, just google for it) but the disc.
then I because a citizen and the government controlled me.
Point.
Now I'm a consumer, and all my rights are under control.
Excercise the right not to buy. Didn't that thought ever occur to anybody?
Making rip-proof media is fine, if they can figure out how to do it. You are agreeing to a contract in regards to the content on whatever media you are purchasing. Self-destructing is destroying my property, unless I can re-use the media it came on.
If I don't like what a company is doing, I will not buy their goods. I still don't buy Colgate products because of them bullying ajax.org.
Stop pretending like you don't have rights. You have rights, you chose not to use them. You are a consumer, and you will buy everything they tell you to buy. Then you bitch when they take you all for suckers.
If you don't like it, don't buy it. If you do buy it, don't bitch.
Don't get me wroing, I don't love AIX by any stretch of the imagination ;-) But this is starting to seem like the technology equivalent of Days of our Lives or something!
AIX is great, if it's used for what it's designed for... but that's not why I'm responding.
I see this as the equivalent to Celebrity Deathmatch, except those episodes where it is a stale, crusty has-been going against a massive gorilla.
Think of Ozzy vs. Mike Tyson, and chuckle.
Yeah, right, and people should be writing in Esperanto on their Dvorak layout keyboards, too, but that's about the same likelihood.
Est tajp jam nun, kun Dvorak... sed mi.
Well, libSDL is what throws the decoded PNG onto the screen. You do need both; one decodes, the other displays the decoded result.
Or you could just learn XLib and not deal with an entire external library... I think I addressed this in my post previously.
>>Do I _REALLY_ want to pull in libpng and libSDL just to do this?
>Yes.
Uhm, to display a PNG to the screen. I can understand pulling libpng, but not libSDL.
It sounds like you either have a delusionally high estimation of your own abilities relative to those of your library-writing peers ("Of course the code I write will be faster, more correct, and less bloaty than the specialised code that those people who really understand the problem and spend time solving it well!") or you place very little value on your time. Oh well. Have fun writing png-frobbing code for the rest of your life.
Most open source code is shit. It's not because there aren't great programmers, it's because there are a shitload of bad programmers. That whole 80/20 thing. Unless you are familiar with a library, I don't blame anybody for not using it for small projects.
If I was writing a png viewer, I'd do it with libpng and X. X isn't that hard, it's a pain in the ass to work with for large programs, but it's not that hard.
Big Deal. Hotel rooms should basically have a nice bed and nice bath. Personally, if I am going on vacation, I want to spend as much time OUT of the hotel room as possible but when I come back I want to get a good nights sleep and get clean the next morning. This type of stuff is designed to attract the people who bought those old Acer computers just because they had a black case.
You are completely and totally wrong. There are tons of reasons why this is a good thing. Executive travelling, where they aren't on vacation but on a business trip. Second, visiting friends on vacation to a place where you know the area, and don't need to site see but just hang out with your friends.
This type of stuff is designed to attract people who have money, and want to stay in comfort.
Just because you don't understand something, doesn't mean it's bad/stupid/wrong. It just means you don't understand the need for it.
I guess you also don't understand why people like to own a BMW, or Mercedes.
North America and Europe were the only two continents industrialized enough to need that amount of oil.
You seem to be forgetting those little events called World War.
So, we showed them it was there (all they saw was a vast, useless desert), gave them the technology to extract it, and gave them the markets to sell their otherwise useless oil to.
As opposed to be conquered by Germany, and having it stolen and their people raped and murdered... I can see why they hate the US and Britain...
It's unlikely that more than 1-2% of US desktops are running a Free operating system.
It was unlikely in 1995. You are forgetting that there has been Lindows, huge marketing pushes, and tons of geeks doing installfests. I would say it's closer to 5%.
As for poor quality software, I suppose you haven't used BIND or Sendmail, eh? Even "better" software (Apache, Samba, OpenSSH, etc.) still has remote root holes not too uncommonly, and the Linux kernel has had hundreds of local root holes.
This is just a dumb argument, nobody calls their own child ugly.
What kind of language model did you use for the IRC analyser? It would probably possible to create a model resilient to bad spelling.
Just statistics based upon word relations, ignoring certain prepositions. I was testing an idea that I had to determine if you could find the difference between languages (like SVO/SOV syntaxes) through statistical analysis.
You can also check my JE, I post in there quite a bit.
I'm not sure it will work very well. In fact I doubt it will, information extraction works best for huge corpuses of text with some fairly simple regularity in their structure. It can still be natural language, but one might make the observation that A is usually before B, and bias the system for such a special case.
Just for an algorithm idea, if you scan a huge glob of text (web-pages, etc) and form patterns based upon word relevance and relation, and then categorize the webpage (manually or through automated tools... screenscarping google, etc.) and have a final result with loose attachments to category. This is just for building the statistics engine, however. I was working on an idea for this using IRC text, but considering most of the people on IRC type with the acumen of autistic lepars, it didn't turn out to well.
There was one project where they mined medical journal articles for which genes were found to affect which properties. I don't have a link to the paper right now but if you're interested I can look it up.
Ironic... after my NN work I went to work writing genetic sequence analysers.. Mostly finding patterns in the contigs to find A/T stops (much more challenging than it seems, especially when a 0% false-positive rate has to be there)
I might put up a website at some point. The project is both school work and commercial work so I'll get around doing it at some point. The interesting part is school work and the useful part (the application) is commercial.
We should talk more, if you have the time/desire I am pretty sure I'd love to bring you into nerdfarm. Just a general mish-mash of open source work, some of it is theoretical. We're currently building the website framework right now so there isn't anything to see on the page.
Perhaps you should just familiarize yourself a bit more with UNIX command line utilities; the ease and power with which skilled UNIX users can find and organize things is one of the major selling points of UNIX.
Uhm, typical "I'm a l33t l33n0x" response. Thanks for wasting your time on an attempt to try to make yourself seem more elite. If you had actually read what I was going for, you would understand that your system of symlinking and fgrep lines makes you look like a fucking retard. Like this:
Organizing data is hard, but chances are that whatever you want to do is already pretty easy with existing tools in UNIX/Linux; otherwise, people would have already added tools for addressing the problem long ago.
Because Haystack is doing something different than what is done, and Mac OSX is working towards the same goal. Linux just clones what the competition is doing, a few years later.
Thanks for playing, save your elitism for high school.
The fact that neither of these terms have shown up in the discussion thus far indicate how little business training most slashbots have. See, sometimes, there's value to an MBA. If nothing else, we're more conversent with lawyer-ese than most programmers.
;) I didn't realize you were an MBA... I could definitely use some of your help if you are interested.
I knew there were the terms, but couldn't remember what they were. Thanks
(And in case you were wondering what all this means, basically, anyone who downloaded the software might have it free and clear. AOL, via Nullsoft, via authority of Justin, may have to abide by the original terms, GPL.)
That was what I was getting at. If the deal that AOL has over Nullsoft allows Nullsoft to act and release software independantly of the AOL label (such as Winamp, which afaik, is still Nullsoft Winamp, not AOL Winamp) than Justin could authorize the release under the GPL (excluding the mess with RSAREF.)
Is that correct?
Bayesian filtering is another of those words that mean a lot of things ;)
Isn't it just...
A widely used set of language models are the Hidden Markov Models. I'm planning to use them on an information extraction problem (populating database tables from free-text descriptions), and that's about the closest to the problem we're discussing that I've been. You could probably use them as a partial solution here as well, but I can't think of any really clever scheme at the moment.
Do you have a website or anything for this? It seems pretty cool, we should talk. If you need webspace or a server for it I can provide it.
And finally, sorry to be boring, but I'm not currently working on anything that would create something like the system we've discussed =)
I'm enjoying the conversation, I wish I had more time to dedicate to it but I am working on something... that isn't like what we're discussing at all, and is ultimately boring. As the saying goes, "Character is doing something that needs done with the interest in doing it fades away." At least we get to put in some cool magic in the backend.
What I would like to have would be something that could integrate all information related to some "project" into a single, powerful interface. Something that would allow me to have information in different formats (txt, html, pdf, ps, doc, spreadsheets, images and so on) searchable from the same locations. Also it would need to have a convenient way of "commenting" material written by others and linking these commments to the right place in the right document, without me having to place e.g. HTML links in the right locations. Then discussion material (mailing-lists, forums) should also be integrated in the same interface.
For me, this would be the ultimate window manager. When I'm on a computer, I'm generally working... or playing games but then I'm on Windows... so anyway. Having a truly integrated desktop would be awesome, but unfortunately too messy (especially with Linux) because there is no standard. If people were to actually come up with and adhere to a standard application model, something like this would be possible. Until that day, you'd have to re-invent so many wheels it wouldn't be feasible.
I guess my main point is that different media should not be artificially classified by its media type but rather by its contents (or more specifically which project it belongs to). My secondary point would then be that everything should be searchable, rather than found using a URL. More like the web and a relational database and less like a filesystem tree.
Exactly - and that's what I would like. Even if I have to at first "train it" what things look like.
In regards to plain text/personal information -- have you thought about looking at Bayesian filtering for a solution to that? I haven't (yet) but the idea is festering in my brain.
is extended ext2 attributes (google for the kernel patch and userland utils)
It allows you to give attributes to files, (like "category")
I understand there's also gnome support.
Index the attributes and put limited statistical information checks on it (also access count, would be nice) and yeah... pretty much.
I might misunderstand you, but does this not mean that the "AI"-methods used by Haystack would not be AI?
r y
Using AI methods does not constitute AI. I haven't looked at their design papers so I can't say for sure. I have yet to see a real AI implementation, so I would say they use AI-Methods, but not AI.
I also have some Neural Network / Statistical processing background so I tend to share this same view that the intelligence is in the designer not in the program. However, the term AI is often used to describe both neural and statistical methods.
Most people view AI as some ambigious concept that they can't understand, so they just slap the label AI on anything that involves automated decision making... *shrug*
Actually neural networks could also be perceived as statistical methods. I don't know about fuzzy logic but my intuition tends hint that they arent fundamentally very different either, just based on different mathematical theory.
Fuzzy Logic is massively statistical, just a bunch of true-false statistics.
To return to your relational filesystem idea, I would prefer having both relational search capabilities and google-like utilization of implicit information. I don't know about you, but I certainly am not very good at being consistent, so search and selection methods allowing some kind of fuzziness criteria would certainly be nice. The basic OS could still work with files, but my personal information should be organized in some more practical way, and should preferably be available all the time from diverse clients (home computer, work computer, PDA, cellphone...).
I'm consistent by putting them in directories... that's about as far as it goes. I try to organize it like $DEVROOT/(Commercial|Contract|OpenSource)/$Catego
That never really goes through... being able to
store everything and pull it up using different methods would kick ass. Sometimes I write a perl script for Project "Oops, I forgot" that would be great in my current project. Digging through all of my source takes ages, but having something to figure out A) It's a perl script (easy), B) roughly what it's purpose is (hard) would be awesome.