Err, no, phones for us don't exist. As far as I can see, there are only two types of phones: phones that are designed to be as cheap as possible, which you apparently think are made with us in mind (they aren't), and phones that are bloated.
I don't want the irrelevant features, but I do want relevant features, and I do want a small and well built phone with a long battery life. Where is a reliable, quad-band GSM phone that doesn't have all of the features unrelated to being a phone? And does it have an efficient user interface?
I do care about battery life. That "better screen" requires a backlight to be able to see anything on it, whereas older screens didn't. The faster processors and memory required to deal with video and other such things don't help.
I also care about interface efficiency for doing what a phone is supposed to do. Every phone I have had has had a less efficient interface than the last one, as keys become used for irrelevant functions. The latest phone I worked with required more button presses to get to the recent calls list than to get to the ringtone settings. Some of the keys were configurable of course, but none of the available options had anything to do with the phone functions. Apparently, the calculator, planner, WAP browser, ringtone settings, games, and so on, were all considered more important by the UI designers than a list of people who I called or or called me recently, and whom I might want to call again!
I also care about reliability. I've been using my latest cell phone for the last 4 years or so. Everyone I talk to who has a newer phone seems to go through at least one a year.
I would gladly pay 500 or 600 dollars for a small, reliable, and well built phone with a black and white display, quad band GSM, an efficient UI, no bloat (only phone, modem, and possibly SMS functions), a long battery life, and a three year warranty. Water resistance or proofing would be a nice plus. Unfortunately, most cell phone manufacturers seem to think that one either wants the cheapest phone possible, or wants an extremely bloated phone which does everything well except calls.
I will have to take back my previous comments about there being no official talk about selling the machines commercially. The OLPCWiki discusses it here. Apparently, commercial sales as a subsidy are being considered, at around 3x the cost of the laptop (so around $400 to $500, probably).
There seems to be some contradictory information on that. The OLPC Wiki's market FAQ seems to imply that the laptops will belong to the children, at least after they finish school.
Just go to a few conventions that used to be Linux strongholds and look at how many powerbooks and macbooks are there now.
Be sure to check whether those are running OS X or Linux. I had an Ibook for a few years that ran OS X for only a few hours every few months (when I wanted to try it out). The hardware is sometimes worth it (Of course, now I have a Dell laptop which came with free 3 year next business day warranty support for higher education. I don't need reliable hardware if I can have someone come replace everything every few months. I'm already on my third motherboard, third keyboard, and second hinge, and the new motherboard is defective too.)
I said many of us end up compiling some things from source. I've had to do that for niche applications on Fedora, Suse, and slackware.
I had thought you were talking about autoconf only from the context. This is a problem. The hope is that most home users won't need those niche applications. But there is definitely room for improvement there.
Most major software is provided in multiple package formats for the different distributions. Lots of minor software provides one and the source. The source can be troublesome and the packages take extra work and cannot always be shared between distributions. So I have to download a different one for each distro I'm running.
I doubt that that is going to be solved unless one distribution gains significant marketshare, unfortunately.
Which is still not as easy as dragging and dropping or throwing something in the trash. Say you have two laptops running the same distro and you want to copy a program from one to the other. You already have a directory shared. According to the desktop metaphor you should be able to just drag the application from one computer to the other or even just run it off the mounted directory. It is intuitive, but it does not work on Linux unless someone is running GNUStep and happens to have something packaged for it, which is rare in the extreme. What about binary software that goes away. I've IM'd copies of such software to friends simply by dragging the icon into my IM chat with them. That is easy and intuitive, but it does not work on Linux.
That is quite interesting. It should be possible with some work to set up a system in Ubuntu that would act like that using deb packages and a modified file manager. It would also probably not be a bad idea to have themes set up that way, as opposed to the current, very inflexible theme manager. I will have to discuss it with some others. Unfortunately, it would only work with Ubuntu...
By which I believe you mean, most software does not adhere to a standard but sends random network traffic often in obfuscated formats. As security advances we'll have our programs running in jails and be able to restrict them at will from accessing any given resources. Having an official service for updates and registration in a clear XML format makes it clear what is going out from what application and to whom and lets the user restrict network access for say, a FPS video game from sending out network traffic once it has been registered and until you want to look for an update and lets a user examine what is being sent. This is vital if we ever want to mitigate trojans. It also lets us build standard tools to manage updates and installs, updating everything at, say 1am, instead of when you run it next and want to get work done.
A standardised system for updates for third-party programs would be great. Ubuntu is already able to update automatically at set times, if I recall, but only for software in the repositories. As for standardised registration, do you mean registration as in network-based licensing, or registration as in registering with a server? A standard for the latter would be nice, but I doubt that developers requiring the former would be willing to use a standard system.
At least on #ubuntu, experienced users/zealots usually tell users asking about how to install a specific software to use apt-get on the command line because, when giving instructions, it is the easiest way. For Ubuntu, there is the Add/Remove Programs app, which end users are supposed to use. But what is easier, telling a user to open a terminal from the menu and type in sudo apt-get install program, or telling the user to open Add/Remove Programs, type the name of the program, check the checkbox next to it, and click Install? While the latter might be the most intuitive, the former is far superior in a support situation, especially since there is little room for confusion, and the instructions are far easier to follow, even if it doesn't make sense.
Sharpmusique worked well for buying music off of Itunes the last time I tried it, and had a quite intuitive installation in Ubuntu (download.deb, double click, type in password, press Install).
Perhaps you are using the wrong distributions? Most major linux distributions are not like Gentoo.
Except that the plan isn't to make them ubiquitous, the plan is to only give them to students, and not allow others to buy them, even at a significantly higher price. There are at least several thousand people in the US who would pay $300 or more for one of the laptops (I would probably pay $600 or so if it were sold officially).
The deal has not, as far as I am aware, been discussed as a possibility officially. It is only a proposal made by unrelated people, and due to the wording of the pledge (signatories pledge to only purchase the laptop for $300, and only if over 100000 people sign by October), it is inconceivable that it will be successful except as an indication that there is interest in buying the laptops at an increased price.
Also, from the prices I have seen, $300 barely pays for two computers. It certainly won't pay for three - the laptops cost significantly more than $100 to build.
It would be a far better to discuss the possibility of buying the laptops at a greatly increased price (probably more around $500 or $600) directly with the project. If a show of support is necessary, then a petition on much more general terms would be far more useful (something like "I would be interested in buying one at a greatly increased price.").
That may give you a hint as to why people are moving from Linux to OS X in droves for workstations.
Would you care to provide some significant non-circumstantial evidence for that rather strong assertion, or are you just taking into account the small number of people whom you happened to meet?
Installing software only from a terminal is fine for a server, or even a developer workstation, but it is not ideal for a normal user's desktop or workstation. Compare the instructions you listed above to dragging an application anywhere on your hard drive, which is what OS X provides. That may give you a hint as to why people are moving from Linux to OS X in droves for workstations. The other part is that such an installation does not always work on every linux variant without dealing with numerous other factors. It does not always handle dependencies, nor updates, nor licensing, nor registration. It does not handle permissions on a per user and per system basis. It does not allow already installed programs to be portable. A standard that does not provide the benefits and features needed to compete is largely useless.
What distributions have you been using where you need to compile everything from source? Ubuntu's package manager installs just about everything, and many third party groups are starting to provide compatible.debs as well. For that, using gdebi, synaptic, and the unknown program in the menu (Add/Remove Programs), it is simple to install both packages from the repositories and external packages. In most external cases it is as simple as double clicking on the deb, or clicking on it in a browser, and then pressing Install on the dialog that comes up. Gdebi also manages dependencies for the packages. As for add/remove programs, installing and removing software is just a matter of manipulating checkboxes. We could work on third-party updates, which is lacking, but as far as I know, most programs that do this do it on their own (firefox, etc...). Licensing and registration are the same - most programs I have seen deal with that on their own.
As for single packages like OS X, I have separate partitions for/home,/usr, and/var. How is a package going to deal with that without being installed? In many cases the separation isn't trivial -/usr may be on a flash drive, or even read only.
Wikipedia has a reasonably comprehensive article on the term and its origins here.
Personally, I believe that a better joke at Senator Stevens' expense could be had by noting that this might significantly speed up the transmission of the internets that his staff sends him.
Actually, if he had read the bill carefully, he would have seen that the bill does not allow for unblocking of sites with educational value. It allows for the blocking to be disabledby request for a minor with adult supervision for educational purposes. This is quite different from not blocking sites with educational value.
The point is that one could make a difference, but not with this pledge. Why does the wording not only include the 100,000 disclaimer, but also add "only if" to this? This is like adding "I don't support this proposal if it doesn't get over 75% of voters in favour" to a vote. It is quite possible that far fewer than 100,000 people would be enough to convince the OLPC to sell the laptops in this way, but this restriction makes the pledges much less clear. Furthermore, the pledge is restricted to $200, which, from the cost estimates I have seen, is less than the cost of two of the laptops. Why not just have a pledge that says something like "I would buy one for a significantly higher price if available."?
A better way of making a difference, in my opinion, would be to send letters to the OLPC Project, instead of just signing up to the flawed pledge. This presents the argument more directly to the people who can actually make this decision.
That pledge seems doomed to failure. Having 97000 more people sign up before the end of October is inconceivable - it would require around 3 people signing up every 4 minutes for the next 90 days. The petition is worded in a way such that if there are not 100000 signatures, none of the pledges are valid: nothing less than a fully successful drive can be submitted for consideration. Even then, it would only be a suggestion. Articles on the laptops have repeatedly stressed that these will not be for sale, and I don't recall anything that indicates something like this is even being thought about by the association itself.
Unfortunately, due to the demand for them in demographics such as the Slashdot readership, this probably means that there will be a somewhat lucrative market for laptops taken from or sold by the students. The colour and other such anti-theft features won't prevent them from being desireable to people who actually want the specific laptop. While there will certainly be abuse in these sales, if a student can sell their laptop for $400 and then claim to have lost it, using the proceeds from the sale to purchase a replacement, then the only ones hurt by the sale will be those who subsidise the production. It may be that OLPC will be compelled to sell to the general public for this reason.
Using./-filename instead of -filename should work in these cases.
Re:HFS++ looking pretty sharp now eh?
on
WinFS Gets the Axe
·
· Score: 1
The file changed feature does not need to be a part of the filesystem's driver. IMon, for example, is independent of the filesystem used, so long as the filesystem is local.
"No taxation without representation" is a rallying cry, not a law. There are a variety of cases where taxpayers are unable to vote, such as in the cases of the District of Columbia, felons, and minors. I, for example, payed taxes on earned income for years as a minor without being able to vote (for representatives who would push to make it so that my job in a biotech lab wasn't technically illegal!).
From the Google Suggest FAQ, "Google Suggest uses data about the overall popularity of various searches to help rank the refinements it offers." Perhaps they have turned off suggestions for a few obvious terms (porn, etc), but I doubt they are actually filtering the searches used to rank the refinements - this would be quite difficult to do in general. Expecting google to weed through all of the searches that have been made in order to find what some might consider to be illegal would be absurd.
That is mostly how the GPL and most other free software licenses work, actually, and seems to be a rather common misconception. The owner of the software can choose to only release it to certain people, under a license like the GPL. The catch is that then those people can also choose, if they wish, to release the software to others, under the same license (at least in the case of the GPL). While I don't know of any cases where is this actually done outside of a organization, it can be useful for software that is only meant to be distributed inside a company.
Free Software doesn't mean you have to give your software to everyone who wants it
Here is a better example, consisting of me randomly pressing my fingers into the keyboard while pressing space and enter every once in a while. The following text, when copied three times, will give a 97% chance of being an authentic scientific paper. Again there is a curve here, with two reps giving an 80% chance and four giving 91%:
hsflhakjdfhaksehnioanevoiralewytuakeltvkaseln
vasodvalskdhtnaksdltaesoiutylvnaesytaesntrvaestyav es
tvsdatysavdiutlvamsiudtylvnaseytuivalsetnvasuletya s
yetysaudtievnlasyetuklvnaysetuvnlashtjkvnlasdya
tsentvausielntvyaskeultnvyaweiulatvyqpevbytknxcalt va
stydviuanesyltuqawletgveashuotipoasdntvuaestlnvasy ukt
ast yausiety asieun ltyeiuslnvytaukesl btyeu
t
aewyt ueity uieltnyeuakltfaksdfl e
i reoi aehiutale 8tu4 l5yuaiw ltyaesiul tyas8eu thea
t hasekt eahst uiaes't ysea8s]
t5y2243oa theio
thaset a
es t
ae taeios t.sze tliae thaew
taehio
taeht aleht eshtiua;e hta.e thkajethau ethauile htu aelht jkaeth ae
it heaui; theasu; thaeskjl thaesk
tahe ukt
e thauoe fgyu8aes tuaeio thaet n eu alernvyaeuls ey
ra vyeusalv ryekulvanetyiqewuo5238v yreiul aye
ra eyr7eaiob y583ow;q vhteksal hfeiua oyre
av ryeiu layr3iuqlr54hcj3q2kbg5cuieap ry3iwulqv ryeuiaslcre
areyiurlv aeruclaneyril3bqvtr7eoavn uer yea' rae
r eatuiraevbryaeiuaprv6a;e'e're yesi pryawe8
ryweaiu rla2tl2gjrv geui fyfeoi tlh4e4oyu r0hu trshgae
tehSUK GFASUD TYUEIA LTUKL HIU lguksdl yr8t; eYiul sety
a tguail tGUIESl teuseli thh us'ty art
aet u
ae teuatsU It89aw3 tuiae tha"R tASREtaesiuo t;asegt
as gt iutgae7i3p ly0qtu seoiatg hak. h5owei utg8eahg
hatuske thakuslt hu3ail 5g3uil3w gt7iao seyt
aety aiusel tgAUIlr tyai7l wytoaw
t e3yesual egtke.sag kj,egUISLEyr7 iulye3t5ou'
er tyeuisl YTEiulty34uialthdjkgldhsgiuLYGUei ha
dihfakjsdlhf ueliYTuisdlfyg73ilhf U
EFheskjlghaesu krhuIL yukdlfhgasdkjgh' GUDSKLA YU
YTUAISELTHJ" UTJSDLHT KLHkj lhtueil HTAESU
HSUetly udlatye7la hgdkjgl hsdktlat h.
Read the paper listed in the menu of the website. The system essentially compresses the text with different window sizes, and then looks at the compression factors. In other words, it is only looking for repetition of strings. This is absurdly easy to fool, and the MIT generator could be easily fixed to pass this filter. For example, try entering a random text once (your post, for example). Note that it fails. Then append a few copies of the same text, and run that through. Your post, when run once, is too short. When run with two copies, it is rejected as 41.2%. When run with three, it passes with 93%. There is a window of repetition level required in order to pass - papers that do not repeat enough are classified as fake, as well as papers that repeat too much (try entering twenty copies of your post).
It should be relatively simple to make a random paper generator that always passes this test with a higher probability than human-written papers.
Err, what Fry's stores do you go to? At the ones in San Diego, they close the boxes up, stick little "this item was returned" stickers on them, and put them back on the shelves again, usually for the same price, or with a nominal discount (~5%). This is another thing to look out for when shopping at Fry's.
Slashdot Mirror
Err, no, phones for us don't exist. As far as I can see, there are only two types of phones: phones that are designed to be as cheap as possible, which you apparently think are made with us in mind (they aren't), and phones that are bloated.
I don't want the irrelevant features, but I do want relevant features, and I do want a small and well built phone with a long battery life. Where is a reliable, quad-band GSM phone that doesn't have all of the features unrelated to being a phone? And does it have an efficient user interface?
I don't care about expense.
I do care about battery life. That "better screen" requires a backlight to be able to see anything on it, whereas older screens didn't. The faster processors and memory required to deal with video and other such things don't help.
I also care about interface efficiency for doing what a phone is supposed to do. Every phone I have had has had a less efficient interface than the last one, as keys become used for irrelevant functions. The latest phone I worked with required more button presses to get to the recent calls list than to get to the ringtone settings. Some of the keys were configurable of course, but none of the available options had anything to do with the phone functions. Apparently, the calculator, planner, WAP browser, ringtone settings, games, and so on, were all considered more important by the UI designers than a list of people who I called or or called me recently, and whom I might want to call again!
I also care about reliability. I've been using my latest cell phone for the last 4 years or so. Everyone I talk to who has a newer phone seems to go through at least one a year.
I would gladly pay 500 or 600 dollars for a small, reliable, and well built phone with a black and white display, quad band GSM, an efficient UI, no bloat (only phone, modem, and possibly SMS functions), a long battery life, and a three year warranty. Water resistance or proofing would be a nice plus. Unfortunately, most cell phone manufacturers seem to think that one either wants the cheapest phone possible, or wants an extremely bloated phone which does everything well except calls.
If I recall correctly, the indirect petroleum usage from the farming is quite significant, is it not?
I will have to take back my previous comments about there being no official talk about selling the machines commercially. The OLPCWiki discusses it here. Apparently, commercial sales as a subsidy are being considered, at around 3x the cost of the laptop (so around $400 to $500, probably).
There seems to be some contradictory information on that. The OLPC Wiki's market FAQ seems to imply that the laptops will belong to the children, at least after they finish school.
Be sure to check whether those are running OS X or Linux. I had an Ibook for a few years that ran OS X for only a few hours every few months (when I wanted to try it out). The hardware is sometimes worth it (Of course, now I have a Dell laptop which came with free 3 year next business day warranty support for higher education. I don't need reliable hardware if I can have someone come replace everything every few months. I'm already on my third motherboard, third keyboard, and second hinge, and the new motherboard is defective too.)
I had thought you were talking about autoconf only from the context. This is a problem. The hope is that most home users won't need those niche applications. But there is definitely room for improvement there.
I doubt that that is going to be solved unless one distribution gains significant marketshare, unfortunately.
That is quite interesting. It should be possible with some work to set up a system in Ubuntu that would act like that using deb packages and a modified file manager. It would also probably not be a bad idea to have themes set up that way, as opposed to the current, very inflexible theme manager. I will have to discuss it with some others. Unfortunately, it would only work with Ubuntu...
A standardised system for updates for third-party programs would be great. Ubuntu is already able to update automatically at set times, if I recall, but only for software in the repositories. As for standardised registration, do you mean registration as in network-based licensing, or registration as in registering with a server? A standard for the latter would be nice, but I doubt that developers requiring the former would be willing to use a standard system.
At least on #ubuntu, experienced users/zealots usually tell users asking about how to install a specific software to use apt-get on the command line because, when giving instructions, it is the easiest way. For Ubuntu, there is the Add/Remove Programs app, which end users are supposed to use. But what is easier, telling a user to open a terminal from the menu and type in sudo apt-get install program, or telling the user to open Add/Remove Programs, type the name of the program, check the checkbox next to it, and click Install? While the latter might be the most intuitive, the former is far superior in a support situation, especially since there is little room for confusion, and the instructions are far easier to follow, even if it doesn't make sense.
.deb, double click, type in password, press Install).
Sharpmusique worked well for buying music off of Itunes the last time I tried it, and had a quite intuitive installation in Ubuntu (download
Perhaps you are using the wrong distributions? Most major linux distributions are not like Gentoo.
Except that the plan isn't to make them ubiquitous, the plan is to only give them to students, and not allow others to buy them, even at a significantly higher price. There are at least several thousand people in the US who would pay $300 or more for one of the laptops (I would probably pay $600 or so if it were sold officially).
The deal has not, as far as I am aware, been discussed as a possibility officially. It is only a proposal made by unrelated people, and due to the wording of the pledge (signatories pledge to only purchase the laptop for $300, and only if over 100000 people sign by October), it is inconceivable that it will be successful except as an indication that there is interest in buying the laptops at an increased price.
Also, from the prices I have seen, $300 barely pays for two computers. It certainly won't pay for three - the laptops cost significantly more than $100 to build.
It would be a far better to discuss the possibility of buying the laptops at a greatly increased price (probably more around $500 or $600) directly with the project. If a show of support is necessary, then a petition on much more general terms would be far more useful (something like "I would be interested in buying one at a greatly increased price.").
Would you care to provide some significant non-circumstantial evidence for that rather strong assertion, or are you just taking into account the small number of people whom you happened to meet?
What distributions have you been using where you need to compile everything from source? Ubuntu's package manager installs just about everything, and many third party groups are starting to provide compatible .debs as well. For that, using gdebi, synaptic, and the unknown program in the menu (Add/Remove Programs), it is simple to install both packages from the repositories and external packages. In most external cases it is as simple as double clicking on the deb, or clicking on it in a browser, and then pressing Install on the dialog that comes up. Gdebi also manages dependencies for the packages. As for add/remove programs, installing and removing software is just a matter of manipulating checkboxes. We could work on third-party updates, which is lacking, but as far as I know, most programs that do this do it on their own (firefox, etc...). Licensing and registration are the same - most programs I have seen deal with that on their own.
As for single packages like OS X, I have separate partitions for /home, /usr, and /var. How is a package going to deal with that without being installed? In many cases the separation isn't trivial - /usr may be on a flash drive, or even read only.
Surely you are joking here?
Wikipedia has a reasonably comprehensive article on the term and its origins here.
Personally, I believe that a better joke at Senator Stevens' expense could be had by noting that this might significantly speed up the transmission of the internets that his staff sends him.
Actually, if he had read the bill carefully, he would have seen that the bill does not allow for unblocking of sites with educational value. It allows for the blocking to be disabled by request for a minor with adult supervision for educational purposes. This is quite different from not blocking sites with educational value.
The point is that one could make a difference, but not with this pledge. Why does the wording not only include the 100,000 disclaimer, but also add "only if" to this? This is like adding "I don't support this proposal if it doesn't get over 75% of voters in favour" to a vote. It is quite possible that far fewer than 100,000 people would be enough to convince the OLPC to sell the laptops in this way, but this restriction makes the pledges much less clear. Furthermore, the pledge is restricted to $200, which, from the cost estimates I have seen, is less than the cost of two of the laptops. Why not just have a pledge that says something like "I would buy one for a significantly higher price if available."?
A better way of making a difference, in my opinion, would be to send letters to the OLPC Project, instead of just signing up to the flawed pledge. This presents the argument more directly to the people who can actually make this decision.
That pledge seems doomed to failure. Having 97000 more people sign up before the end of October is inconceivable - it would require around 3 people signing up every 4 minutes for the next 90 days. The petition is worded in a way such that if there are not 100000 signatures, none of the pledges are valid: nothing less than a fully successful drive can be submitted for consideration. Even then, it would only be a suggestion. Articles on the laptops have repeatedly stressed that these will not be for sale, and I don't recall anything that indicates something like this is even being thought about by the association itself.
Unfortunately, due to the demand for them in demographics such as the Slashdot readership, this probably means that there will be a somewhat lucrative market for laptops taken from or sold by the students. The colour and other such anti-theft features won't prevent them from being desireable to people who actually want the specific laptop. While there will certainly be abuse in these sales, if a student can sell their laptop for $400 and then claim to have lost it, using the proceeds from the sale to purchase a replacement, then the only ones hurt by the sale will be those who subsidise the production. It may be that OLPC will be compelled to sell to the general public for this reason.
Using ./-filename instead of -filename should work in these cases.
The file changed feature does not need to be a part of the filesystem's driver. IMon, for example, is independent of the filesystem used, so long as the filesystem is local.
"No taxation without representation" is a rallying cry, not a law. There are a variety of cases where taxpayers are unable to vote, such as in the cases of the District of Columbia, felons, and minors. I, for example, payed taxes on earned income for years as a minor without being able to vote (for representatives who would push to make it so that my job in a biotech lab wasn't technically illegal!).
Why do you think that we wouldn't all die if it was caused by Nature?
A quick search turns up the FAQ here. Perhaps they moved it and didn't update the link?
From the Google Suggest FAQ, "Google Suggest uses data about the overall popularity of various searches to help rank the refinements it offers." Perhaps they have turned off suggestions for a few obvious terms (porn, etc), but I doubt they are actually filtering the searches used to rank the refinements - this would be quite difficult to do in general. Expecting google to weed through all of the searches that have been made in order to find what some might consider to be illegal would be absurd.
That is mostly how the GPL and most other free software licenses work, actually, and seems to be a rather common misconception. The owner of the software can choose to only release it to certain people, under a license like the GPL. The catch is that then those people can also choose, if they wish, to release the software to others, under the same license (at least in the case of the GPL). While I don't know of any cases where is this actually done outside of a organization, it can be useful for software that is only meant to be distributed inside a company.
Free Software doesn't mean you have to give your software to everyone who wants it
Here is a better example, consisting of me randomly pressing my fingers into the keyboard while pressing space and enter every once in a while. The following text, when copied three times, will give a 97% chance of being an authentic scientific paper. Again there is a curve here, with two reps giving an 80% chance and four giving 91%:
hsflhakjdfhaksehnioanevoiralewytuakeltvkaseln vasodvalskdhtnaksdltaesoiutylvnaesytaesntrvaestyaRead the paper listed in the menu of the website. The system essentially compresses the text with different window sizes, and then looks at the compression factors. In other words, it is only looking for repetition of strings. This is absurdly easy to fool, and the MIT generator could be easily fixed to pass this filter. For example, try entering a random text once (your post, for example). Note that it fails. Then append a few copies of the same text, and run that through. Your post, when run once, is too short. When run with two copies, it is rejected as 41.2%. When run with three, it passes with 93%. There is a window of repetition level required in order to pass - papers that do not repeat enough are classified as fake, as well as papers that repeat too much (try entering twenty copies of your post).
It should be relatively simple to make a random paper generator that always passes this test with a higher probability than human-written papers.
Err, what Fry's stores do you go to? At the ones in San Diego, they close the boxes up, stick little "this item was returned" stickers on them, and put them back on the shelves again, usually for the same price, or with a nominal discount (~5%). This is another thing to look out for when shopping at Fry's.