I hate web interfaces. I'm happy they've gotten better, and also realize that in many instances, there are no alternative, but in general, I don't think the web browser was made for such type of interactions, nor should it be relied on for such hardships.
While complex apps have been written for the web, not to mention the great things google has done, they are in the end still very hack-oriented.
I think what is really needed to get things off the ground as far as usability is something that is: (a) (really) cross-platformable, (b) language-agnoistic, (c) easy to run/use/create, (d) fast, (e) easy to use with most server types.
While by now I'm sure many people are ready to flame me that Java answers all of these, regardless of how you feel about it, it has not been the success that was once predicted. As long as the majority of internet applications are using the web for interfaces, it is not a success.
Many of the pieces are in place for such a thing. For example, Mozilla's framework may work, or perhaps the Trolltech toolkit QT, or maybe even Java. But with all of these there are still big obstacles. Can you make an interface in 5-10 minutes? How difficult is it to set up the server to handle the requests? Can you run the server easily anywhere (or will the cheaper hosting company only allow HTTP requests?).
Perhaps the best approach, is to continue using HTTPXMLRequests for data transmission, as that is the status quo, but for someone to come up with a complete solution to quickly build such interfaces. I think Interface Builder for Cocoa is a great goal for how one might put together such an app. Tie that together with something like Glade, or Renessance, it would make for a easy to allow for different languages/toolkits to use, and then have a standard API/library for communicating with the server. People can write bindings to their favourite languages, and you have something that is mostly cross-platformable. The only caveat, is for the scripting language, a correct interpreter is required per platform. One solution, would be to allow automatic download of the correct interpret for your platform if it is not detected.
At least in the good old days, Google was in the business of selling their search engine. Originally their website was ad-free, and was used to prove how much better their engine was than everyone else (now that they have adsense, it is in some ways not so different how ID has both a business making games, and selling their engine).
If you go to their Business Solutions, you will see that while Adsense is listed first, after that they have their Google Search Appliances, and Google Web Search engine.
While they were a private company, they kept very secretive about their revenue stream, but now they are a publically traded company, you can google for more info.
A big advantage to the.xxx domain might not come from a strict legal mandate that pr0n sites must exist there, but rather from accountability. That is, by using that domain name, a company would be free from prosecution for distributing material that someone might find objectionable.
At least in the US, it is up to a community to decide what is acceptable material to display in public. Because of its global nature, this standard cannot and should not be used on the internet. What community is there to decide?
While many people on this site (me included) would like to have a place of completely free communication, we do have to face the fact that many people (most importantly, most politicians) disagree with this point of view. To be realistic, some compromise is needed.
The.xxx may afford such a compromise. Perhaps it would be more paltable if instead we named it '.free', but the idea is the same. If you post something under that domain name, no one should be able to argue (although I'm sure some lawyer in the end will anyways) that the material is indecent. By typing those three letters, you are already acknowledging your intentions.
I would be against *requiring* companies to use '.xxx', and thus filtering would not be a fool-proof. What I would suggest instead of black-listing sites with a domain name, is to have a white-list domain name (maybe '.fam'?). In the same logic before, sites that have domain names there would have to agree to certain standards (perhaps there could be whole families of such domain names, so that different standards could be set).
In the forum comments, yes (technically I wouldn't consider it TFA..)
Assuming the windows port doesn't have any other major depencies (e.g. to play video, etc.) it should be fairly easy to port to Linux. Especially considering (at least at one point [I'm assuming]) it was compiled under gcc.
And I would hazard a guess most competent programmers with some Linux experience should be able to take on such a project.
While this doesn't answer your code directly, depending on what toolkits you use etc. will depend on how easy it is to make a Linux port. For example, if you use a large amount of Cocoa-only library, it will end up being a large amount of work. Some libraries (remember Loki?) particularly may be easy to use on Mac and Windows platforms, but not on more free platforms.
This is in part why the SDL was created. Linked with openGL and openAL, it should be much less laborious for a Linux hacker to get your game working.
Can you determine how involved a port is? The API of your game is well-abstracted from the actual platform, then finding someone capable of porting the game should be much easier than someone who isn't.
Finally, some of the more obvious places to look for people capable of porting said game, is to look through some opensourced projects. This has the advantage that you can get a sense of what someone is capable of doing, being able to see the person's source code, etc.
What I have wanted for a while, is website where people could comment on the best program to use if you want to do X. It doesn't have to have a *single* best answer, but provide some information, and allow a default setting for people who might feel overwhelmed with choices.
The website would have two modes: consumer, and reviewer. The consumer mode would provide a tree-like interface where you could click choices on what you want to do. A couple of examples would be: draw->raster->most powerful draw->vector->simple interface files->format->mk2fs publish->math->LaTeX
The reviewer mode would consist of entering a program, and filing it under the catagories you think best. As most projects are moving targets, it would set your comments to a specific version. Based on many people's input, the website would determine which projects each project would most likely fall into (so it can occur in several catagories at once). This also allows a consumer to come along, type in a program she/he uses, and find similiar programs that might fit his/her needs better.
Many systems have way more programs installed than an average (or expert) user knows about. There might very well be a program that can fit your needs, but without proper document/direction, can be difficult to find. Thus, a super-directory of all programs available would be a huge boost to the OSS community.
I recently also purchased one. I'm still trying to decide whether it was a good idea, or if I should try ebaying it. It's true, it is cheap. However, you need to count in the fact that you *really* need to get an SD card ($60), and possibly an CF card ($20) if you want to put any other apps on it. At least the instructions I read, said you could only update the ROMs with a CF card, but you want the SD card, because the WIFI card ($35) you will want to put into it will also want to go in the CF slot.
The biggest complaint against these that I read before my purchase is battery life. I wanted something that I easily sync (think rsync or unison) with my computers, and do a limited amount of things (ssh is great, etc.) So I wasn't too worried about battery life. The standard amount of time quoted is about 3hrs, and that is what I found. Until I put the WIFI card in (also Linksys, btw). Then it drops to under an hour. Not that useful anymore. It might make sense to get a PDA with WIFI built in for energy-consumption..
I researched PDAs for a bit before my purchase. After experience with a palm, I swore off all PDAs. It was way too slow for me to enter in data with graffiti, or other such systems. I type fast, and I wanted to be able to enter data quickly into the system. True, they have keyboards for PDAs, but (a) they are expensive, and (b) it's a lot less portable now. With the Zaurus, although it has a small keyboard, I can type fairly quickly for most things. If PDAs are ever going to work, this is certainly the best direction.
My conclusion so far. All PDAs suck. This one sucks the least for its price. I like the look of the ArchOS PMA400, but once again, no keyboard, no purchase.
I'm probably posting too late for most people to read, but it seems weird to me that the two original scientists made the discovery, and believe that it works, why not go into the electrical power industry? (of course, I also ask the same question of the crackpots who make perpetual motion machines, though this does not reflect on my stance on whether cold fusion)
If the material is easy to find, and equipment cheap, then they should be able to produce cheap power with little overhead. From my understanding, it's pretty easy to sell power back to the electrical company, so a very small scale start-up would be possible as a first step.
For obvious reasons, the GPU will be best at doing linear algerbra related problems quickly. I am curious, however, whether anyone has tried numerical integration (I've gone through several sites before without seeing mention of this). While the GPU may not be the best suited for this, it seems like it might be possible to still offload some work done by the CPU with the GPU, and potentially speed up simulations.
If anyone has any information on this, I would be very interested to hear (or perhaps other hardware solutions for solving ODEs quickly)
In line with what KDE and GNOME do (e.g. `gconf-config --libs-dirs`), why not have a single program that reports where different things are supposed to go? This would save the difficulty of having to having these companies/orgs actually agreeing to things, and would make it easy to make sure things always go in the right place (e.g. a makefile can simply do 'sys-config --install-bin-dir' to figure out where to install the resulting binaries). You don't even need to get the distros to agree, as these things can be fairly easily maintained by a third party. All you need to do is make sure this program always goes in the same location (e.g./sbin/sys-config). Might even be able to replace autoconfig/automake by letting the program advertise the capabilities of the system (i.e. programs can register/unregister capabilities).
Perhaps the best way to fight spam is a public campaign about not responding to it. And if that doesn't work, threats on people's lives could also work. Or maybe blame it on making children starve to death in some remote country. I suspect there is some critical mass that is required to keep it going.
X11R6 is the protocol used, not the implementation. The XFree86 project was a popular project to implement the X11 protocol on 8086 machines. Because of politics, the X client/server was branched by x.org (as the source code was open), so a different development style could be done (and from what I understand many of the people from the XFree86 project switched to x.org as well). There do exist other implementations of X11 that cost money, but at least in the past advertised better performance.
Most people intelligent people I know do not adhere strictly to either the democrat's or republican's policies, but rather their views fit in better with either a smaller party, or none of the above.
However, when it comes to voting it is only one of those two parties who usually receive any significant votes. That said, most libertarians I know vote republican.
At first glance, the reason for this is fairly obvious. Republicans say that they will keep government out of your life by reducing taxes, and keeping the government smaller. What I have not gotten answered by my libertarian friends is why companies are necessarily so much better.
That is, I think it's a given someone in the end is going to try to run our lives. Be it RIAA, MPAA, doubleclick, Time-Warner, etc. While a free market is a great idea, we seem to have difficulty in its actual practice. Our media outlets have shrunk, only a few companies control needed supplies (e.g. ADM, who was found guilty in price-fixing), etc.
If neither companies nor government can be completely trusted, it seems better to me to instead try to balance the two. How does this fit in the the libertarian view point (i.e. companies and government should be of equal mass)? If freedom is the focus of your party, then it is important to specify from whom. This may give a more liberal-bent to the libertarian party as well.
While I'm sure you rather people vote for you, if someone were bent on voting republican or democrat, but with libertarian values, do you think this should influence the decision?
Not if you use a sane encryption scheme. With your logic, all our encrypted data will be hackable in a couple years, and thus not very useful. The biggest difficulty is getting cheap hardware that can do the decrypt quickly.
I am curious as to how they will manage encryption with this, and if it will be yet another encryption through obfuscation.
It seems the smartest approach is to publish and patent the encryption scheme, but make it so time consuming, that you will need hardware to do the decryption properly. That way any one who tries to get around the protection scheme and not pay royalties will be easily sueable.
The upside for non-mainstream OS users, is that it will most likely mean non-OS dependent solutions (maybe).
Of course programmable logic chips could potentially be a threat, but not a major one, as most people don't have that type of hardware.
I used to really be into PHP. It was great for creating a webpage in little or no time. It's syntax, while maybe not perfect, was pretty good. Until, that is, I tried to start developing my own libraries, and ran into weird quirkiness with object design, and trying to figure out the best way to do libraries, etc. It looks like PHP5 might fix the problems I had with how objects worked, and I'm not sure if it was my own fault with the messy libraries I ended up with, or whether I didn't find the best way to do it in PHP, but I eventually moved to Python out of frustration.
I had avoided Python for a long time, as I really disliked (and still do) the indentation-matters issue. But besides that, and its own set of quirks, it's a really great language, and for larger projects I have trouble even thinking about going back to PHP.
I think the biggest selling point to PHP over other solutions such as Python is that its simple. You don't have to make a whole of choices. For example, with Python you have a large number of packages to choose from: Zope, mod_python, twisted.web, Python CGI, and a bunch of variants on these. While choice can be good, it can also be overwhelming (like how do you know which package to go with until you've tried them all?).
I think I am not alone with some of difficulties I faced with PHP. So while it's great to hear that PHP has fixed many of its bugs, I think its worthwhile for people to also look at other solutions out there.
(1) Most search engines have to be made general to appeal to the public, and without much personal information about you (which is probably a good thing). Enter search engine filtering via the browser. In conjunction with a preexisting search engine (e.g. google), filter out the results based on human feedback, and previously tagged websites. You can allow different 'modes' to enhance the search abilities. For example, I might have a mode to search for Neuroscience articles. When in this mode it could filter based off of rules it learned when I browse the web in that mode.
Another addition might be page ranking.
(2) Intelligent page-merging. Given N number of pages, determine what information is related among those pages, and try to only include the most informative. For example, I might be trying to learn more on the mean-value theorem, and it would then take bits and pieces of several pages to form a larger text. As the internet is full of half-written articles on subjects, this could be potentially very useful.
(3) I think various versions of this might have been posted, but: personal library. As you browse the web, instead of having to deal with bookmarking web pages, it would be really nice if the browser could sort those webpages into something intelligent. For example, if I were looking up information on Python, it could collect those pages into some type of tree, so when I want to go back to look something up, I could just navigate that tree. Probably would be nice to be able to turn this feature off too.
(4) Alternative formatting. Although there exists standards for how to best format webpages so agents go automatically go through and collect information, many web pages do not use this. For humans too, organization of web pages can often be problematic. It would be cool if the web browser could automatically organize the web page into topics/subtopics/etc. This could allow pages to be browsed more like books.
(5) For Thunderbird, it would be cool to have 'topics'. That is, have the email client automatically catagorize emails into different topics, and be able to view the email based on that (extra points for graphical representations). For example, suppose I wanted to recall all the emails I had sent concerning ACh (acetylcholine). Thunderbird would have to first figure out exactly what that is, figure out how it related to my other emails (e.g. neurotransmitters), to allow for a hierarchical structure of topics. Thus I could set my topic to either neurotransmitters or ACh, and see all the corresponding emails. While this can mostly be emulated by searches on 'subject' and 'body', by allowing a hierarchical structure you would get emails that did not contain those specific keywords.
Slashdot Mirror
While complex apps have been written for the web, not to mention the great things google has done, they are in the end still very hack-oriented.
I think what is really needed to get things off the ground as far as usability is something that is: (a) (really) cross-platformable, (b) language-agnoistic, (c) easy to run/use/create, (d) fast, (e) easy to use with most server types.
While by now I'm sure many people are ready to flame me that Java answers all of these, regardless of how you feel about it, it has not been the success that was once predicted. As long as the majority of internet applications are using the web for interfaces, it is not a success.
Many of the pieces are in place for such a thing. For example, Mozilla's framework may work, or perhaps the Trolltech toolkit QT, or maybe even Java. But with all of these there are still big obstacles. Can you make an interface in 5-10 minutes? How difficult is it to set up the server to handle the requests? Can you run the server easily anywhere (or will the cheaper hosting company only allow HTTP requests?).
Perhaps the best approach, is to continue using HTTPXMLRequests for data transmission, as that is the status quo, but for someone to come up with a complete solution to quickly build such interfaces. I think Interface Builder for Cocoa is a great goal for how one might put together such an app. Tie that together with something like Glade, or Renessance, it would make for a easy to allow for different languages/toolkits to use, and then have a standard API/library for communicating with the server. People can write bindings to their favourite languages, and you have something that is mostly cross-platformable. The only caveat, is for the scripting language, a correct interpreter is required per platform. One solution, would be to allow automatic download of the correct interpret for your platform if it is not detected.
If you go to their Business Solutions, you will see that while Adsense is listed first, after that they have their Google Search Appliances, and Google Web Search engine.
While they were a private company, they kept very secretive about their revenue stream, but now they are a publically traded company, you can google for more info.
How's this then:
The color will start off as a nice #FF00FF, but later on the day is expected to go a #F400A1.
Actually the current threat level of the worm is light fusia. However, experts are predicting it might go to dark fusia by tonight.
All j00r base are belong to us!
That's "humanity's" sake initwit!
A big advantage to the .xxx domain might not come from a strict legal mandate that pr0n sites must exist there, but rather from accountability. That is, by using that domain name, a company would be free from prosecution for distributing material that someone might find objectionable.
.xxx may afford such a compromise. Perhaps it would be more paltable if instead we named it '.free', but the idea is the same. If you post something under that domain name, no one should be able to argue (although I'm sure some lawyer in the end will anyways) that the material is indecent. By typing those three letters, you are already acknowledging your intentions.
At least in the US, it is up to a community to decide what is acceptable material to display in public. Because of its global nature, this standard cannot and should not be used on the internet. What community is there to decide?
While many people on this site (me included) would like to have a place of completely free communication, we do have to face the fact that many people (most importantly, most politicians) disagree with this point of view. To be realistic, some compromise is needed.
The
I would be against *requiring* companies to use '.xxx', and thus filtering would not be a fool-proof. What I would suggest instead of black-listing sites with a domain name, is to have a white-list domain name (maybe '.fam'?). In the same logic before, sites that have domain names there would have to agree to certain standards (perhaps there could be whole families of such domain names, so that different standards could be set).
For humanities sake, PLEASE stop the iJokes!
Monk-boy: There is no linux-kernel.
In the forum comments, yes (technically I wouldn't consider it TFA..)
Assuming the windows port doesn't have any other major depencies (e.g. to play video, etc.) it should be fairly easy to port to Linux. Especially considering (at least at one point [I'm assuming]) it was compiled under gcc.
And I would hazard a guess most competent programmers with some Linux experience should be able to take on such a project.
While this doesn't answer your code directly, depending on what toolkits you use etc. will depend on how easy it is to make a Linux port. For example, if you use a large amount of Cocoa-only library, it will end up being a large amount of work. Some libraries (remember Loki?) particularly may be easy to use on Mac and Windows platforms, but not on more free platforms.
This is in part why the SDL was created. Linked with openGL and openAL, it should be much less laborious for a Linux hacker to get your game working.
Can you determine how involved a port is? The API of your game is well-abstracted from the actual platform, then finding someone capable of porting the game should be much easier than someone who isn't.
Finally, some of the more obvious places to look for people capable of porting said game, is to look through some opensourced projects. This has the advantage that you can get a sense of what someone is capable of doing, being able to see the person's source code, etc.
What I have wanted for a while, is website where people could comment on the best program to use if you want to do X. It doesn't have to have a *single* best answer, but provide some information, and allow a default setting for people who might feel overwhelmed with choices.
The website would have two modes: consumer, and reviewer. The consumer mode would provide a tree-like interface where you could click choices on what you want to do. A couple of examples would be:
draw->raster->most powerful
draw->vector->simple interface
files->format->mk2fs
publish->math->LaTeX
The reviewer mode would consist of entering a program, and filing it under the catagories you think best. As most projects are moving targets, it would set your comments to a specific version. Based on many people's input, the website would determine which projects each project would most likely fall into (so it can occur in several catagories at once). This also allows a consumer to come along, type in a program she/he uses, and find similiar programs that might fit his/her needs better.
Many systems have way more programs installed than an average (or expert) user knows about. There might very well be a program that can fit your needs, but without proper document/direction, can be difficult to find. Thus, a super-directory of all programs available would be a huge boost to the OSS community.
I recently also purchased one. I'm still trying to decide whether it was a good idea, or if I should try ebaying it. It's true, it is cheap. However, you need to count in the fact that you *really* need to get an SD card ($60), and possibly an CF card ($20) if you want to put any other apps on it. At least the instructions I read, said you could only update the ROMs with a CF card, but you want the SD card, because the WIFI card ($35) you will want to put into it will also want to go in the CF slot.
The biggest complaint against these that I read before my purchase is battery life. I wanted something that I easily sync (think rsync or unison) with my computers, and do a limited amount of things (ssh is great, etc.) So I wasn't too worried about battery life. The standard amount of time quoted is about 3hrs, and that is what I found. Until I put the WIFI card in (also Linksys, btw). Then it drops to under an hour. Not that useful anymore. It might make sense to get a PDA with WIFI built in for energy-consumption..
I researched PDAs for a bit before my purchase. After experience with a palm, I swore off all PDAs. It was way too slow for me to enter in data with graffiti, or other such systems. I type fast, and I wanted to be able to enter data quickly into the system. True, they have keyboards for PDAs, but (a) they are expensive, and (b) it's a lot less portable now. With the Zaurus, although it has a small keyboard, I can type fairly quickly for most things. If PDAs are ever going to work, this is certainly the best direction.
My conclusion so far. All PDAs suck. This one sucks the least for its price. I like the look of the ArchOS PMA400, but once again, no keyboard, no purchase.
Perhaps part of the problem is people downloading their favourite infected app..
I'm probably posting too late for most people to read, but it seems weird to me that the two original scientists made the discovery, and believe that it works, why not go into the electrical power industry? (of course, I also ask the same question of the crackpots who make perpetual motion machines, though this does not reflect on my stance on whether cold fusion)
If the material is easy to find, and equipment cheap, then they should be able to produce cheap power with little overhead. From my understanding, it's pretty easy to sell power back to the electrical company, so a very small scale start-up would be possible as a first step.
For obvious reasons, the GPU will be best at doing linear algerbra related problems quickly. I am curious, however, whether anyone has tried numerical integration (I've gone through several sites before without seeing mention of this). While the GPU may not be the best suited for this, it seems like it might be possible to still offload some work done by the CPU with the GPU, and potentially speed up simulations.
If anyone has any information on this, I would be very interested to hear (or perhaps other hardware solutions for solving ODEs quickly)
In line with what KDE and GNOME do (e.g. `gconf-config --libs-dirs`), why not have a single program that reports where different things are supposed to go? This would save the difficulty of having to having these companies/orgs actually agreeing to things, and would make it easy to make sure things always go in the right place (e.g. a makefile can simply do 'sys-config --install-bin-dir' to figure out where to install the resulting binaries). You don't even need to get the distros to agree, as these things can be fairly easily maintained by a third party. All you need to do is make sure this program always goes in the same location (e.g. /sbin/sys-config). Might even be able to replace autoconfig/automake by letting the program advertise the capabilities of the system (i.e. programs can register/unregister capabilities).
Just a thought...
Perhaps the best way to fight spam is a public campaign about not responding to it. And if that doesn't work, threats on people's lives could also work. Or maybe blame it on making children starve to death in some remote country. I suspect there is some critical mass that is required to keep it going.
we can start to use the 'back in my days, our laptops were made of wood, AND we had to walk up hill both ways...'
X11R6 is the protocol used, not the implementation. The XFree86 project was a popular project to implement the X11 protocol on 8086 machines. Because of politics, the X client/server was branched by x.org (as the source code was open), so a different development style could be done (and from what I understand many of the people from the XFree86 project switched to x.org as well). There do exist other implementations of X11 that cost money, but at least in the past advertised better performance.
Most people intelligent people I know do not adhere strictly to either the democrat's or republican's policies, but rather their views fit in better with either a smaller party, or none of the above.
However, when it comes to voting it is only one of those two parties who usually receive any significant votes. That said, most libertarians I know vote republican.
At first glance, the reason for this is fairly obvious. Republicans say that they will keep government out of your life by reducing taxes, and keeping the government smaller. What I have not gotten answered by my libertarian friends is why companies are necessarily so much better.
That is, I think it's a given someone in the end is going to try to run our lives. Be it RIAA, MPAA, doubleclick, Time-Warner, etc. While a free market is a great idea, we seem to have difficulty in its actual practice. Our media outlets have shrunk, only a few companies control needed supplies (e.g. ADM, who was found guilty in price-fixing), etc.
If neither companies nor government can be completely trusted, it seems better to me to instead try to balance the two. How does this fit in the the libertarian view point (i.e. companies and government should be of equal mass)? If freedom is the focus of your party, then it is important to specify from whom. This may give a more liberal-bent to the libertarian party as well.
While I'm sure you rather people vote for you, if someone were bent on voting republican or democrat, but with libertarian values, do you think this should influence the decision?
Not if you use a sane encryption scheme. With your logic, all our encrypted data will be hackable in a couple years, and thus not very useful. The biggest difficulty is getting cheap hardware that can do the decrypt quickly.
I am curious as to how they will manage encryption with this, and if it will be yet another encryption through obfuscation.
It seems the smartest approach is to publish and patent the encryption scheme, but make it so time consuming, that you will need hardware to do the decryption properly. That way any one who tries to get around the protection scheme and not pay royalties will be easily sueable.
The upside for non-mainstream OS users, is that it will most likely mean non-OS dependent solutions (maybe).
Of course programmable logic chips could potentially be a threat, but not a major one, as most people don't have that type of hardware.
I used to really be into PHP. It was great for creating a webpage in little or no time. It's syntax, while maybe not perfect, was pretty good. Until, that is, I tried to start developing my own libraries, and ran into weird quirkiness with object design, and trying to figure out the best way to do libraries, etc. It looks like PHP5 might fix the problems I had with how objects worked, and I'm not sure if it was my own fault with the messy libraries I ended up with, or whether I didn't find the best way to do it in PHP, but I eventually moved to Python out of frustration.
I had avoided Python for a long time, as I really disliked (and still do) the indentation-matters issue. But besides that, and its own set of quirks, it's a really great language, and for larger projects I have trouble even thinking about going back to PHP.
I think the biggest selling point to PHP over other solutions such as Python is that its simple. You don't have to make a whole of choices. For example, with Python you have a large number of packages to choose from: Zope, mod_python, twisted.web, Python CGI, and a bunch of variants on these. While choice can be good, it can also be overwhelming (like how do you know which package to go with until you've tried them all?).
I think I am not alone with some of difficulties I faced with PHP. So while it's great to hear that PHP has fixed many of its bugs, I think its worthwhile for people to also look at other solutions out there.
Just my $0.02.
(1) Most search engines have to be made general to appeal to the public, and without much personal information about you (which is probably a good thing). Enter search engine filtering via the browser. In conjunction with a preexisting search engine (e.g. google), filter out the results based on human feedback, and previously tagged websites. You can allow different 'modes' to enhance the search abilities. For example, I might have a mode to search for Neuroscience articles. When in this mode it could filter based off of rules it learned when I browse the web in that mode.
Another addition might be page ranking.
(2) Intelligent page-merging. Given N number of pages, determine what information is related among those pages, and try to only include the most informative. For example, I might be trying to learn more on the mean-value theorem, and it would then take bits and pieces of several pages to form a larger text. As the internet is full of half-written articles on subjects, this could be potentially very useful.
(3) I think various versions of this might have been posted, but: personal library. As you browse the web, instead of having to deal with bookmarking web pages, it would be really nice if the browser could sort those webpages into something intelligent. For example, if I were looking up information on Python, it could collect those pages into some type of tree, so when I want to go back to look something up, I could just navigate that tree. Probably would be nice to be able to turn this feature off too.
(4) Alternative formatting. Although there exists standards for how to best format webpages so agents go automatically go through and collect information, many web pages do not use this. For humans too, organization of web pages can often be problematic. It would be cool if the web browser could automatically organize the web page into topics/subtopics/etc. This could allow pages to be browsed more like books.
(5) For Thunderbird, it would be cool to have 'topics'. That is, have the email client automatically catagorize emails into different topics, and be able to view the email based on that (extra points for graphical representations). For example, suppose I wanted to recall all the emails I had sent concerning ACh (acetylcholine). Thunderbird would have to first figure out exactly what that is, figure out how it related to my other emails (e.g. neurotransmitters), to allow for a hierarchical structure of topics. Thus I could set my topic to either neurotransmitters or ACh, and see all the corresponding emails. While this can mostly be emulated by searches on 'subject' and 'body', by allowing a hierarchical structure you would get emails that did not contain those specific keywords.