Redhat 9 seems to be doing some of this, just having an install me a desktop system option, and having items on the 'start' menu say things like 'Instant messenger' instead of gaim - (It doesnt go so far as to set it up for you unfortunately, one of the most unintuitive things I found was having to add a plugin just to use msn).
Two things I suspect it still needs though - firstly, the added help function for things like - 'On windows I used word- HELP!!!' which would point you to openoffice.
The second would definitely be having some form of media player in the default install (mplayer or xine) that could play video (including dvd, mpeg4/divx, quicktime, realplayer) and not just xmms. When I installed redhat for a friend they found it really easy to use, but the first thing they wanted to do was watch a movie, which is easier said than done. With windows, theres a media player built in. Aside from that though, I found redhats default install to have pretty much anything needed for someone just migrating from windows who want to do office work, browse the web, etc... etc..
Now make 3 of these, and with the remaining plastic bit, you can make a borg cube to annoy your nice little fleet. Its still not a patch on putting cd's in the microwave and making pieces of art out of the interestingly patterned results.
(I would make a comment about shoving the enterprise in the microwave and ion storms here, but thats going a bit too far)
I know the planet in the picture isnt far enough away in this case, but it would be great if hubble showed us a picture of a proto-planet/star/galaxy a few million light years away, why by now, has become an earth like planet, with an advanced civilisation that has discovered FTL travel, who are now on their merry way to earth to say hi! (Or less great - with big guns aimed in our direction).
Thats great, when you want to look at new sites, or only want to keep track of a few. When you want to look up a lot of the same sites regularly any method other than bookmarks/open in tabs simply doesnt cut it imho. E.g. I have one set of tabs for all of the news sites/forums I read - just fire up open in tabs and post away, or much more valuable, all the web comic sites I read daily (userfriendly, garfield, dilber, sinfest, megatokyo etc..) are in a sinlge bookmark folder, which saves me having to type in ~10 different urls (and growing fast) for what amounts to about 30 seconds of reading funnies.
As for keeping bookmarks in sync, I still havent solved that problem satisfactorily - thankfully I only use two different profiles atm, so I just copy the bookmarks.html file over from one computer to another whenever I modify it. Web based bookmarks would be nice, but you lost the open in tabs feature, and its much nicer having everything on a little toolbar rather than having to wait for a website to load.
The only thing I could think of would be a script that reguarly checks a central location for the bookmarks file, and updates it cvs style when changes are made locally (cvs itself seems a little overkill but you get the idea). One alternative I am not sure is possible would be to store your profile online, this would require you use the same browser across the board (I do), but could concevably slow everything down each time it access the profile across the net.
It is actually possible to have a firmware update increase the speed of your drive. A friend of mine did it (albeit accidentally) when flashing their dvd drive to be region free (It was a 12x, they used a rom for the 16x version of the drive). The problem with this however, is that the 12x drive obviously wasnt designed to be run faster than that (or presumably it would already) and problems could occur with the extra heat generated. I dont thik the drive lasted too long.
OTOH, I wouldnt be surprised if there are drives out there that are sold like processors, where some models tested to be capable of higher speeds during QA are sold as lower speed drives through demand.
The think is, it wouldnt make much difference anyway, and the prices of most drives are that cheap anyway it wouodnt matter (okay, DVD writing drives are still pretty expensive, and unfortunately, in this case you would be doubling the write speed of the drive, which I am fairly sure would cause overheating problems, not to mention wearing out parts quicker).
Unfortunately, partly due to the main use of p2p being to share content that some people dont want you to have without paying (call it pirated material if you will, but I digress), some of us are pretty much forced to limit our upstream traffic, as it it carefully monitored to detect any use of p2p tools results in immediate disconnection. (Somehow I doubt they would differentiate between using bittorrent to get the latest redhat, and using it to get the latest movie - the terms state p2p is prohibited because it causes network saturation (allowing would mean everyone downloading/uplading gigs/day in the eyes of those who make the rules), and servers in general are prohibited (it is a university network, and the intent is to prevent the like of DHCP servers taking everything out, but in practise it can apply to anything)).
Granted, bittorrent is a great solution, and I have used it many times on occasion, but I don't think it should replace anoymous ftp servers.
Then again, I'm running debian, so I guess it doesnt really matter.:)
I for one wasn't, but to quote an example where I believe this hasnt applied - removal of copy protection for the purposes of compatibility (playing dvds on linux).
If the above example is wrong in some way, then would that provision allow the development of an open source driver for SD cards for the sharp zaurus SL-5500 for use in openzaurus, assuming of course that it was reverse engineered and not obtained by other means? Because this would be specifically for the purposes of compatibility, and something in which I hold great interest.
I'd warrant that a lot of the terms of this law that apply to computers and are actually valid (i.e. not arreasting somebody for ssh chaining withing their lan or something ridiculous), are actually already enforcable.
To use the example of using NAT when the provider charges per computer, this will be spelled out in the contract, and therfore the company would be within their rights to sue you for breach of contract, and most likely have criminal charges brought against you for fraud.
While I'm on the subject of now allowing NAT on the network (which my current provider does - for mostly valid reasons - the intent is to prevent one person in the halls of residence paying, and the others freeloading off the same connection), I have a main computer, and a headless, openBSD box to act as a firewall/NAT, I also have a networked printer (connected via ethernet, and accessible through port forwarding remotely), and a handheld with ethernet card. All of these are used by me and there is no intent to screw the university out of money, and yet technically, just browsing the net on my handheld through the main computer is a breach of contract.
disclaimer: I have not (yet) read the paper, just the first linked article.
But the problems you describe wouldnt be problems for say an ad-hoc p2p wireless network, with each node forwarding for others.
- generally, you can't tell what your topoligy your network will end up having, so forget about architecting one
It doesnt really matter, as long as it works and can get packets from a to b, to me it seems the whole point is that you don't need to architect the network.
- it does not guarantee that all your nodes will end up being networked within a fixed number of attempts (see the fig. 3 in the paper)
In a peer-to-peer system, this wouldnt be a big problem, as you state, for a corporate network, this wouldnt be acceptable, but the solution is just to keep trying until you do get networked, or the user gets fed up of waiting.
I guess my point is, given time (your point about it not being established is obviously valid, considering it is just a research paper atm) this could be the solution to the peer to peer wifi network that people have been looking for, automatically finding the best route for traffic, without requiring any centralised control. I would assume they still have to work around the problem of malicious agents leaving strong 'chemical' scents, but it still sounds liek a really neat idea.
One of the big problems with keeping your old number (or at least it used to be) is that a lot of companies have their tarrifs set by number ranges. I remember seeing something for (I think) O2, where you could keep your old number. However, it meant that other people who phoned you from the same network (who would normally get charged very little for calling the same network) now have to pay cross-network charges, which are extortionate to say the least. I cant remember if it worked the other way (i.e. you got charged the expensive rate for calling same network), but it wouldnt surprise me if it did.
And considering I chose my phone provider based on what family and friends mostly used (much like it is with IM systems) to get the cheapest calls, something like this is a pretty big incentive to not keep the same number (or keep the same provider)
link clicks and attachments end up going to mozilla browser windows instead of the preferred browser.
This has been my major gripe with much of the KDE tools (and pretty much any integrated system which simply assumes you must be running all of their tools because you happen to like one). I run phoenix, and getting knewsticker or kmerlin (msn client) to open links in phoenix is pretty much impossible (yes yes.. I know.. use the source luke and all that, but thats time I dont have at the moment - too busy posting to slashdot:P ). And the other example in windows, where any link you click automanically (sic) opens everything in iexplore, despite setting the default browser as phoenix.
A cry out to developers.. please please PLEASE if you have highlighted links in your app, let the user configure which browser they want to start it up in. And not through some weird edit/etc/unituitivename/.hidden_weird_config_file and add the line "faj3fs.kfj.browserN = phoenix", but in the normal settings dialogue. (gaim does this, but there are very few others that I have seen that do).
The users who would benefit most from having binaries (new users who dont know how to compile from source) are unlikely to have non-x86 computers - I would expect a user of a (say) sparc machine to at least know how to type in./configure && make && make install, so if space is really at a premium, provide binaries for just the x86 platform, and have source for everyone else.
Of course, the best solution is user education, perhaps instead of writing 4 pages of how to do an install step, point new users to a central place on how to do it. However, this doesnt help those who simply want to get their computer working and dont want to learn yet another method (e.g. those who migrated from windows systems), which is why we have binary packages that take any work out of the users hands.
I had a similar idea a while back, just with a little more sinister application. It involved using two ultrasound sources with frequencies differering by something like 7Hz (cant remember exactly, but it was supposedly the resonant frequency of the gut), aim it at someone and make them feel distinctly queasy. Thankfully for all my classmates, I never actually got to try it out (I'm not sure if the guy in the article used the same method, and whether my idea would have worked), but if he were to transmit the sound at the natural frequencies of certain objects to that specific object, some pretty spooky effects could be generated, and nobody would hear. (Time to go set up a haunted house I think.. smashing wine glasses, voices in peoples heads, strange feeling in the stomach - 'must be a ghost passing through me'). I'd make a fortune!
The parents subject pretty much says it all. One other issuee is licensing, I was surprised to find that bash wasnt installed by default when I installed OpenBSD, until I realised one of the aims of *BSD is to avoid GPL software. To be honest, the only real difference I noticed between the two was that ksh didnt support my nice luverly coloured bash prompt (which probably says more about how I use the shell than how different they are). However, tab completion worked in ksh, unlike what I took the parent to be suggesting.
Isnt that what package managers are for? Have a developer do the 'hard' work of compiling the program and resolving dependencies, shipping it out as a binary package, and letthing the user simply type apt-get install package (granted with rpm systems the user still has to resolve dependencies manually, but most of the ease is there). Of course, typing any commands in isnt always the most end-user friendly way of doing things.. which explains the existence of the myriad of GUI package management tools.
Personally I like the ports idea of *bsd, donwload a single tar.gz (or use cvs, but we're talking ease of use here), untar it, cd to the correct directory for the package you want to install.. (hmm.. I'm looking to install a networking program, so maybe I should trythe 'net' directory), and just type make install - resolves dependencies, downloads the package, downloads any patches and patches the software for you, configures, compiles, and installs. Presumably gentoo's emerge command works similarly (I havent tried gentoo myself, but anyway).
An user who is just interested in getting the software working and isnt bothered about patching, or having the fastest build for their processor will be perfectly happy with binaries, and if package management tools are still considered too hard, make an installshield style program that leaves one file for the user to run (most linux games I've seen)
Slashdot Mirror
Redhat 9 seems to be doing some of this, just having an install me a desktop system option, and having items on the 'start' menu say things like 'Instant messenger' instead of gaim - (It doesnt go so far as to set it up for you unfortunately, one of the most unintuitive things I found was having to add a plugin just to use msn).
Two things I suspect it still needs though - firstly, the added help function for things like - 'On windows I used word- HELP!!!' which would point you to openoffice.
The second would definitely be having some form of media player in the default install (mplayer or xine) that could play video (including dvd, mpeg4/divx, quicktime, realplayer) and not just xmms. When I installed redhat for a friend they found it really easy to use, but the first thing they wanted to do was watch a movie, which is easier said than done. With windows, theres a media player built in. Aside from that though, I found redhats default install to have pretty much anything needed for someone just migrating from windows who want to do office work, browse the web, etc... etc..
.. How long before someone comes up with a zaurus port?
but this article is giving me severe doubts.
Now I'd have thought that in this case, it is security that is winning out over the ease of use of having older httpd.conf files work.
Now make 3 of these, and with the remaining plastic bit, you can make a borg cube to annoy your nice little fleet.
Its still not a patch on putting cd's in the microwave and making pieces of art out of the interestingly patterned results.
(I would make a comment about shoving the enterprise in the microwave and ion storms here, but thats going a bit too far)
I know the planet in the picture isnt far enough away in this case, but it would be great if hubble showed us a picture of a proto-planet/star/galaxy a few million light years away, why by now, has become an earth like planet, with an advanced civilisation that has discovered FTL travel, who are now on their merry way to earth to say hi! (Or less great - with big guns aimed in our direction).
Thats great, when you want to look at new sites, or only want to keep track of a few.
When you want to look up a lot of the same sites regularly any method other than bookmarks/open in tabs simply doesnt cut it imho. E.g. I have one set of tabs for all of the news sites/forums I read - just fire up open in tabs and post away, or much more valuable, all the web comic sites I read daily (userfriendly, garfield, dilber, sinfest, megatokyo etc..) are in a sinlge bookmark folder, which saves me having to type in ~10 different urls (and growing fast) for what amounts to about 30 seconds of reading funnies.
As for keeping bookmarks in sync, I still havent solved that problem satisfactorily - thankfully I only use two different profiles atm, so I just copy the bookmarks.html file over from one computer to another whenever I modify it. Web based bookmarks would be nice, but you lost the open in tabs feature, and its much nicer having everything on a little toolbar rather than having to wait for a website to load.
The only thing I could think of would be a script that reguarly checks a central location for the bookmarks file, and updates it cvs style when changes are made locally (cvs itself seems a little overkill but you get the idea).
One alternative I am not sure is possible would be to store your profile online, this would require you use the same browser across the board (I do), but could concevably slow everything down each time it access the profile across the net.
It is actually possible to have a firmware update increase the speed of your drive. A friend of mine did it (albeit accidentally) when flashing their dvd drive to be region free (It was a 12x, they used a rom for the 16x version of the drive). The problem with this however, is that the 12x drive obviously wasnt designed to be run faster than that (or presumably it would already) and problems could occur with the extra heat generated. I dont thik the drive lasted too long.
OTOH, I wouldnt be surprised if there are drives out there that are sold like processors, where some models tested to be capable of higher speeds during QA are sold as lower speed drives through demand.
The think is, it wouldnt make much difference anyway, and the prices of most drives are that cheap anyway it wouodnt matter (okay, DVD writing drives are still pretty expensive, and unfortunately, in this case you would be doubling the write speed of the drive, which I am fairly sure would cause overheating problems, not to mention wearing out parts quicker).
its been posted three times now!
Its not just a dupe.. its a tripe!
will somebody please moderate that fossil -1 redundant?
Unfortunately, partly due to the main use of p2p being to share content that some people dont want you to have without paying (call it pirated material if you will, but I digress), some of us are pretty much forced to limit our upstream traffic, as it it carefully monitored to detect any use of p2p tools results in immediate disconnection. (Somehow I doubt they would differentiate between using bittorrent to get the latest redhat, and using it to get the latest movie - the terms state p2p is prohibited because it causes network saturation (allowing would mean everyone downloading/uplading gigs/day in the eyes of those who make the rules), and servers in general are prohibited (it is a university network, and the intent is to prevent the like of DHCP servers taking everything out, but in practise it can apply to anything)).
:)
Granted, bittorrent is a great solution, and I have used it many times on occasion, but I don't think it should replace anoymous ftp servers.
Then again, I'm running debian, so I guess it doesnt really matter.
I for one wasn't, but to quote an example where I believe this hasnt applied - removal of copy protection for the purposes of compatibility (playing dvds on linux).
If the above example is wrong in some way, then would that provision allow the development of an open source driver for SD cards for the sharp zaurus SL-5500 for use in openzaurus, assuming of course that it was reverse engineered and not obtained by other means? Because this would be specifically for the purposes of compatibility, and something in which I hold great interest.
My CRT has topped that one for a while now.
I'd warrant that a lot of the terms of this law that apply to computers and are actually valid (i.e. not arreasting somebody for ssh chaining withing their lan or something ridiculous), are actually already enforcable.
To use the example of using NAT when the provider charges per computer, this will be spelled out in the contract, and therfore the company would be within their rights to sue you for breach of contract, and most likely have criminal charges brought against you for fraud.
While I'm on the subject of now allowing NAT on the network (which my current provider does - for mostly valid reasons - the intent is to prevent one person in the halls of residence paying, and the others freeloading off the same connection), I have a main computer, and a headless, openBSD box to act as a firewall/NAT, I also have a networked printer (connected via ethernet, and accessible through port forwarding remotely), and a handheld with ethernet card. All of these are used by me and there is no intent to screw the university out of money, and yet technically, just browsing the net on my handheld through the main computer is a breach of contract.
Okay.. rant over.. move along
disclaimer: I have not (yet) read the paper, just the first linked article.
But the problems you describe wouldnt be problems for say an ad-hoc p2p wireless network, with each node forwarding for others.
- generally, you can't tell what your topoligy your network will end up having, so forget about architecting one
It doesnt really matter, as long as it works and can get packets from a to b, to me it seems the whole point is that you don't need to architect the network.
- it does not guarantee that all your nodes will end up being networked within a fixed number of attempts (see the fig. 3 in the paper)
In a peer-to-peer system, this wouldnt be a big problem, as you state, for a corporate network, this wouldnt be acceptable, but the solution is just to keep trying until you do get networked, or the user gets fed up of waiting.
I guess my point is, given time (your point about it not being established is obviously valid, considering it is just a research paper atm) this could be the solution to the peer to peer wifi network that people have been looking for, automatically finding the best route for traffic, without requiring any centralised control. I would assume they still have to work around the problem of malicious agents leaving strong 'chemical' scents, but it still sounds liek a really neat idea.
One of the big problems with keeping your old number (or at least it used to be) is that a lot of companies have their tarrifs set by number ranges. I remember seeing something for (I think) O2, where you could keep your old number. However, it meant that other people who phoned you from the same network (who would normally get charged very little for calling the same network) now have to pay cross-network charges, which are extortionate to say the least. I cant remember if it worked the other way (i.e. you got charged the expensive rate for calling same network), but it wouldnt surprise me if it did.
And considering I chose my phone provider based on what family and friends mostly used (much like it is with IM systems) to get the cheapest calls, something like this is a pretty big incentive to not keep the same number (or keep the same provider)
link clicks and attachments end up going to mozilla browser windows instead of the preferred browser.
:P ).
/etc/unituitivename/.hidden_weird_config_file and add the line "faj3fs.kfj.browserN = phoenix", but in the normal settings dialogue. (gaim does this, but there are very few others that I have seen that do).
This has been my major gripe with much of the KDE tools (and pretty much any integrated system which simply assumes you must be running all of their tools because you happen to like one). I run phoenix, and getting knewsticker or kmerlin (msn client) to open links in phoenix is pretty much impossible (yes yes.. I know.. use the source luke and all that, but thats time I dont have at the moment - too busy posting to slashdot
And the other example in windows, where any link you click automanically (sic) opens everything in iexplore, despite setting the default browser as phoenix.
A cry out to developers.. please please PLEASE if you have highlighted links in your app, let the user configure which browser they want to start it up in. And not through some weird edit
Whatever happened to (ogg) tarkin?
The users who would benefit most from having binaries (new users who dont know how to compile from source) are unlikely to have non-x86 computers - I would expect a user of a (say) sparc machine to at least know how to type in ./configure && make && make install, so if space is really at a premium, provide binaries for just the x86 platform, and have source for everyone else.
Of course, the best solution is user education, perhaps instead of writing 4 pages of how to do an install step, point new users to a central place on how to do it. However, this doesnt help those who simply want to get their computer working and dont want to learn yet another method (e.g. those who migrated from windows systems), which is why we have binary packages that take any work out of the users hands.
Games used to be great.. and they dont make them like the used to be.
(reads article to end).. DOH!!
I had a similar idea a while back, just with a little more sinister application. It involved using two ultrasound sources with frequencies differering by something like 7Hz (cant remember exactly, but it was supposedly the resonant frequency of the gut), aim it at someone and make them feel distinctly queasy. Thankfully for all my classmates, I never actually got to try it out (I'm not sure if the guy in the article used the same method, and whether my idea would have worked), but if he were to transmit the sound at the natural frequencies of certain objects to that specific object, some pretty spooky effects could be generated, and nobody would hear.
(Time to go set up a haunted house I think.. smashing wine glasses, voices in peoples heads, strange feeling in the stomach - 'must be a ghost passing through me'). I'd make a fortune!
The parents subject pretty much says it all. One other issuee is licensing, I was surprised to find that bash wasnt installed by default when I installed OpenBSD, until I realised one of the aims of *BSD is to avoid GPL software.
To be honest, the only real difference I noticed between the two was that ksh didnt support my nice luverly coloured bash prompt (which probably says more about how I use the shell than how different they are).
However, tab completion worked in ksh, unlike what I took the parent to be suggesting.
Of course theres always the new adventure shell.
Isnt that what package managers are for? Have a developer do the 'hard' work of compiling the program and resolving dependencies, shipping it out as a binary package, and letthing the user simply type apt-get install package (granted with rpm systems the user still has to resolve dependencies manually, but most of the ease is there).
Of course, typing any commands in isnt always the most end-user friendly way of doing things.. which explains the existence of the myriad of GUI package management tools.
Personally I like the ports idea of *bsd, donwload a single tar.gz (or use cvs, but we're talking ease of use here), untar it, cd to the correct directory for the package you want to install.. (hmm.. I'm looking to install a networking program, so maybe I should trythe 'net' directory), and just type make install - resolves dependencies, downloads the package, downloads any patches and patches the software for you, configures, compiles, and installs. Presumably gentoo's emerge command works similarly (I havent tried gentoo myself, but anyway).
An user who is just interested in getting the software working and isnt bothered about patching, or having the fastest build for their processor will be perfectly happy with binaries, and if package management tools are still considered too hard, make an installshield style program that leaves one file for the user to run (most linux games I've seen)
From the article:
"Stellar Deep is something we really want to do when we have the financial resources to do it"
So I'm guessing its been places on the back burner for a little while, waiting for the right amount of money to come scurrying towards it.
they could market it as a sort of high tech whoopee cushion. Only the victim gets more of a shock than a mere unexpected rude noise.
BZZZZZZTTTTTTTT!!!!