they mis-predict mobile usage
on
WAP Under Fire
·
· Score: 1
I disagree with section 4.1 of "WAP is a Trap -- Reject WAP". In the future, mobile use of the internet will not be restricted to quick, high-urgency stuff. Extensive mobile access will be common, and will take place everywhere people now read books or magazines. This includes resturaunts, parks, busses, waiting rooms, and many other locations that are difficult to wire. The devices used will have high resolution displays, and be capable of high speed compression/decompression. A successful mobile data communications standard must support these devices, because they will generate the majority of the traffic.
Someday, I expect to be able to sit on the porch of any little cafe, order chai tea, open my laptop, and vpn into work.
Good protocols aren't designed to "only last a few years". In part because when the technical limitations they work around are obsolete, the corresponding limitations of the obsolete standard hang around like an albatross.
The license certificate is now a sticker on the PC case. It flakes apart if you try to remove it. This makes it impossible to return the software for refund, without returning the PC too.
Admittedly, only a few people ever got their money back by returning the bundled copy of windows. However, this completely prevents it.
They can triangulate your cell phone's position from the relative signal strength measured at nearby towers. That's accurate to about a hundred meters. Closer than that, they can simply detect your phone as a radio source.
Remember, your phone transmits periodicly, so the cell system knows what tower you're near, to route incoming calls.
Why can't MS's corporate heads be jailed? Bill ``I forgot'' Gates, Ballmer and several others are obviously guilty of perjury and contempt of court. The trick is proving it. However, there's a lot of closed evidence and testamony. Somewhere in all that material there should be something they can be personally nailed on.
I thought that's what ILOVEYOU was. It overwrote a few file types, but really it's warhead was pretty mild. It did just enough to scare people.
It could have done any number of nasty things, for example: email out copies of any files labeled private or confidential, install backorifice and broadcast it's location, erase the flash bios, corrupt wins, corrupt the registry, etc.
Virus scanners are for known viruses. They don't work well on new ones. That's why ILOVEYOU got past them and did so much damage.
Theoretically you could look for "virus-like behavior". I think a lot of scanners can do that, however they usually ship with that disabled. I guess the heuristics aren't good enough yet, and they don't want to annoy people with false hits.
I think you seriously overestimate the intellegence of most outlook users. You seem quite intellegent, but most are idiots. In those conditions "education as an answer" breaks down.
That's our geeky, nerdy, or just non-athletic children who will be falsely reported by the "popular" kids as a new form of high school hazing.
That's our intellegent children who will be falsely reported as dangerous, because they blew the test curve.
The fundamental flaw in this scheme is it asks children to evaluate their peers, yet gives them no incentive to do so fairly. This flaw will cause the system to be abused more than used. It should be shut down before any children are harmed.
Unfortunately, the government will never listen to letters. They don't care about schooling. If they did, public schools would be properly funded, and teachers would be paid decently.
Pinkerton is being paid by the government, probably on a fixed-rate contract. They only care about money, so the cure is to render the system unprofitable. Increase load on the system until it becomes too expensive for them to maintain. I don't mean their network infrastructure. I mean their system of evaluation and screening staff. Trained people are a lot more expensive than equipment.
The true solutions to children's problems in school are the same ones they've always been. Supervision by well trained teachers, backed by good school psychologists. This is currently impossible due to huge class sizes, and poor funding. The money for more teachers should be pulled from quack schemes like Pinkerton's.
I played with some of this stuff. I also tried LiteStep, a NextStep-like replacement for explorer. I never found focus follows mouse. I did find something to map caps-lock to another control. I found that most of the mods sort-of worked in explorer and other pure MS windows things, but broke down in other apps. For example office didn't play nicely with them. Third party stuff was particularly unlikely to work with these alterations. Worse, the stability of the system was seriously reduced.
I complained to a friend who sometimes programs windows apps. His advise was "Windows is inherently broken. Nothing you can do will make it any better."
If there is this much discussion on the issue, then gui design is definately not a completed problem. That means we're still experimenting. Different guis represent diverging research paths. This is good, because it means we're covering more of the problem space.
Bright backgrounds give me eye strain, and headaches. I have to use light text on dark backgrounds.
A while ago I hacked an http proxy to edit out background color codes. That helped a lot. However, SSL and java give me trouble
MS Windows has some support for people with vision problems, but it's incomplete, and the rest of the gui sucks to much to use. X used to have ok support with things like -reverseVideo in the default Xt options list. That worked for Athena Widgets, OpenLook (mostly), and Motif (mostly). However, the newer widget sets are much sloppier. Anything GTK+ based is just horrible. They abandoned all the sophisticated, time tested, functionality of Xt resources and the command line parsing that went with them. Their.gtkrc files provide only rudimentary control, and only from flat files (no xrdb). Worse many of the widgets ignore the gtkrc settings (ie clist). If any of this has changed since 1.2.3, I appologize. However, gtk applications with hard white backgrounds have caused me too many painful headaches for me to be very freindly towards them.
A simple UI, probably copied from windows is only a start. It's ok for casual users, and converting windows users, but it frustrates people who wish to learn and grow into hackers.
A good windowing environment should have simple set of defaults, with a rich set of expert functions available. That way, as curious users become more experianced, they can turn on more things. After they've grown up to be experts, they can still use and like the interface.
A perfect example of a user interface that cannot support experts is ms windows. It's moderately easy to learn. However, there's nothing beyond the beginner mode. People familiar with more sophisticated window managers quickly become frustrated by the lack of common functionality like good icon control, root window menus, fast cut and paste, focus follows mouse, raise-lower, and key bindings.
A good user interface should allow (encourage) the user to "play" with it. They should be able to customize everything. Then they should be able to manage and share those tweaks. In particular, it should be possible to switch between alternate saved configurations on the fly. It should be possible to post those personal configs to the net, and for other people to easily incorporate parts that interest them. Configurations should be commentable. Both simple and complex example configs should be available with the software.
Moreover, it would make it faster for information technology (IT) managers to install computers for new employees since they d no longer be required to copy massive amounts of data each time they set up a new desktop.
This implies a client's files could automaticly become linked to a server's, ie across filesystems. If so, what happens if the client is a laptop and disconnects?
If you ctrl-alt-f2 out of the graphical install to the bash prompt, run fdisk, and go back in, the graphical install doesn't notice you've changed the partition table. There's no refresh button. Even if you go back a few steps and forward again, it doesn't reread the table. It's annoying.
My only complaint with Disk Druid is I have trouble getting it to create the partitions in the correct order. I like to put swap, tmp, and var near the begining of the disk. Rotational delay is smaller there, so you get slightly higher performance. Disk druid seems to like reordering my partitions.
I've used most of the admin tools, linuxconf, smit, sam, etc. They're tolerable for letting an inexperianced sysadmin change something on one box. However, I've never seen one that scaled well to a cluster of machines. When you have 50, 100, or 1000 machines to admin, each with it's own requirements, those tools break down. You have to start scripting.
You also learn it's better to parallel-rdist a config file out, than to try to rsh the same cli admin tool on all boxes. At that point you're back to needing intimate knowledge of the config files. Once you have that knowledge, it's almost always faster to just hack the file than to use the tool.
As for graphical administration and installation tools, please remember many (perhaps most) professionally run unix boxes do not have graphical consoles. They have serial consoles wired into telnet-able terminal servers. The admin may be dialed in from home, or in an office 1000 miles away. Even if the machine is in the same room, a serial console has huge advantages. Cut and paste between the consoles of seperate systems is incredibly useful.
The next step for the software giants would be threatening to sue discussion boards unless they remove negative posts about software. Microsoft could sue/. if someone posted benchmarks for SQLServer. Currently, it's unenforcable (though most of these sites lack the legal resources to oppose a big company). This law would make it enforcable. That would seriously damage our ability to pre-evaluate systems.
They could also try to sue Linus if a driver showed up that they felt was the result of reverse engineering their windows drivers.
When I started doing helpdesk and sysadmin work, users were expected to read the manuals. We would always answer their questions, but we also included a polite reference to the appropriate documentation. After a few examples, we only got calls about the obscure stuff.
I know that sounds harsh, but in the long run it's a kindness to both the user and your tech staff. The users understand things better, we have time to work at improving the system, and everyone's stress level is reduced.
Here's a paraphrase of an old proverb. If you give a man a fish, he eats for a day, then comes back tomorrow to mooch. If you teach him how to fish, he eats for a lifetime.
When did users decide they didn't need to read the docs, not even the brief field descriptions on the screen? More importantly, when did tech support groups decide to coddle stupid lusers with one time answers, instead of teaching them?
Traditionally free software projects have been paid for with peer respect and reputation. Open source generally preserves this. The problem has always been fuguring out how to use peer respect for grocerys. This system has the potential to solve that.
A project sponsor would probably consider the reputation of the bidding developers, and provide qualified ones with sample hardware. For example, almost any ethernet adaptor vendor would probably be willing to give a new card to Donald Becker.
The RIAA's next step would be to lobby for regulations that would require future mp3 players to detect the watermarks, and refuse to play the file. If they couldn't force it with legislation, they would subsidize manufacture of such devices. Their crippled players would be cheaper than the unlimited ones for long enough to drive the unlimited ones off the market.
Slashdot Mirror
I disagree with section 4.1 of "WAP is a Trap -- Reject WAP". In the future, mobile use of the internet will not be restricted to quick, high-urgency stuff. Extensive mobile access will be common, and will take place everywhere people now read books or magazines. This includes resturaunts, parks, busses, waiting rooms, and many other locations that are difficult to wire. The devices used will have high resolution displays, and be capable of high speed compression/decompression. A successful mobile data communications standard must support these devices, because they will generate the majority of the traffic.
Someday, I expect to be able to sit on the porch of any little cafe, order chai tea, open my laptop, and vpn into work.
Good protocols aren't designed to "only last a few years". In part because when the technical limitations they work around are obsolete, the corresponding limitations of the obsolete standard hang around like an albatross.
The license certificate is now a sticker on the PC case. It flakes apart if you try to remove it. This makes it impossible to return the software for refund, without returning the PC too.
Admittedly, only a few people ever got their money back by returning the bundled copy of windows. However, this completely prevents it.
They can triangulate your cell phone's position from the relative signal strength measured at nearby towers. That's accurate to about a hundred meters. Closer than that, they can simply detect your phone as a radio source.
Remember, your phone transmits periodicly, so the cell system knows what tower you're near, to route incoming calls.
Why can't MS's corporate heads be jailed? Bill ``I forgot'' Gates, Ballmer and several others are obviously guilty of perjury and contempt of court. The trick is proving it. However, there's a lot of closed evidence and testamony. Somewhere in all that material there should be something they can be personally nailed on.
#include ``IANAL''
I thought that's what ILOVEYOU was. It overwrote a few file types, but really it's warhead was pretty mild. It did just enough to scare people.
It could have done any number of nasty things, for example: email out copies of any files labeled private or confidential, install backorifice and broadcast it's location, erase the flash bios, corrupt wins, corrupt the registry, etc.
I think that's just NT 3.50, not 3.51 or 4.0. It had to be disconnected from a network. I think it needed some patches too.
Virus scanners are for known viruses. They don't work well on new ones. That's why ILOVEYOU got past them and did so much damage.
Theoretically you could look for "virus-like behavior". I think a lot of scanners can do that, however they usually ship with that disabled. I guess the heuristics aren't good enough yet, and they don't want to annoy people with false hits.
I think you seriously overestimate the intellegence of most outlook users. You seem quite intellegent, but most are idiots. In those conditions "education as an answer" breaks down.
What about doc and xls? They carry most macro viruses.
A lot of the pieces would land on earth. It would probably kill all humans. ... Actually, that does sound rather peaceful...
Most home pcs don't have ethernet cards.
That's our geeky, nerdy, or just non-athletic children who will be falsely reported by the "popular" kids as a new form of high school hazing.
That's our intellegent children who will be falsely reported as dangerous, because they blew the test curve.
The fundamental flaw in this scheme is it asks children to evaluate their peers, yet gives them no incentive to do so fairly. This flaw will cause the system to be abused more than used. It should be shut down before any children are harmed.
Unfortunately, the government will never listen to letters. They don't care about schooling. If they did, public schools would be properly funded, and teachers would be paid decently.
Pinkerton is being paid by the government, probably on a fixed-rate contract. They only care about money, so the cure is to render the system unprofitable. Increase load on the system until it becomes too expensive for them to maintain. I don't mean their network infrastructure. I mean their system of evaluation and screening staff. Trained people are a lot more expensive than equipment.
The true solutions to children's problems in school are the same ones they've always been. Supervision by well trained teachers, backed by good school psychologists. This is currently impossible due to huge class sizes, and poor funding. The money for more teachers should be pulled from quack schemes like Pinkerton's.
I played with some of this stuff. I also tried LiteStep, a NextStep-like replacement for explorer. I never found focus follows mouse. I did find something to map caps-lock to another control. I found that most of the mods sort-of worked in explorer and other pure MS windows things, but broke down in other apps. For example office didn't play nicely with them. Third party stuff was particularly unlikely to work with these alterations. Worse, the stability of the system was seriously reduced.
I complained to a friend who sometimes programs windows apps. His advise was "Windows is inherently broken. Nothing you can do will make it any better."
If there is this much discussion on the issue, then gui design is definately not a completed problem. That means we're still experimenting. Different guis represent diverging research paths. This is good, because it means we're covering more of the problem space.
Bright backgrounds give me eye strain, and headaches. I have to use light text on dark backgrounds.
.gtkrc files provide only rudimentary control, and only from flat files (no xrdb). Worse many of the widgets ignore the gtkrc settings (ie clist). If any of this has changed since 1.2.3, I appologize. However, gtk applications with hard white backgrounds have caused me too many painful headaches for me to be very freindly towards them.
A while ago I hacked an http proxy to edit out background color codes. That helped a lot. However, SSL and java give me trouble
MS Windows has some support for people with vision problems, but it's incomplete, and the rest of the gui sucks to much to use. X used to have ok support with things like -reverseVideo in the default Xt options list. That worked for Athena Widgets, OpenLook (mostly), and Motif (mostly). However, the newer widget sets are much sloppier. Anything GTK+ based is just horrible. They abandoned all the sophisticated, time tested, functionality of Xt resources and the command line parsing that went with them. Their
A simple UI, probably copied from windows is only a start. It's ok for casual users, and converting windows users, but it frustrates people who wish to learn and grow into hackers.
A good windowing environment should have simple set of defaults, with a rich set of expert functions available. That way, as curious users become more experianced, they can turn on more things. After they've grown up to be experts, they can still use and like the interface.
A perfect example of a user interface that cannot support experts is ms windows. It's moderately easy to learn. However, there's nothing beyond the beginner mode. People familiar with more sophisticated window managers quickly become frustrated by the lack of common functionality like good icon control, root window menus, fast cut and paste, focus follows mouse, raise-lower, and key bindings.
A good user interface should allow (encourage) the user to "play" with it. They should be able to customize everything. Then they should be able to manage and share those tweaks. In particular, it should be possible to switch between alternate saved configurations on the fly. It should be possible to post those personal configs to the net, and for other people to easily incorporate parts that interest them. Configurations should be commentable. Both simple and complex example configs should be available with the software.
I want xauth and xhost permissions ANDed together, instead of ORed. So X clients need both a valid cookie, and a valid source ip address.
If you ctrl-alt-f2 out of the graphical install to the bash prompt, run fdisk, and go back in, the graphical install doesn't notice you've changed the partition table. There's no refresh button. Even if you go back a few steps and forward again, it doesn't reread the table. It's annoying.
My only complaint with Disk Druid is I have trouble getting it to create the partitions in the correct order. I like to put swap, tmp, and var near the begining of the disk. Rotational delay is smaller there, so you get slightly higher performance. Disk druid seems to like reordering my partitions.
I've used most of the admin tools, linuxconf, smit, sam, etc. They're tolerable for letting an inexperianced sysadmin change something on one box. However, I've never seen one that scaled well to a cluster of machines. When you have 50, 100, or 1000 machines to admin, each with it's own requirements, those tools break down. You have to start scripting.
You also learn it's better to parallel-rdist a config file out, than to try to rsh the same cli admin tool on all boxes. At that point you're back to needing intimate knowledge of the config files. Once you have that knowledge, it's almost always faster to just hack the file than to use the tool.
As for graphical administration and installation tools, please remember many (perhaps most) professionally run unix boxes do not have graphical consoles. They have serial consoles wired into telnet-able terminal servers. The admin may be dialed in from home, or in an office 1000 miles away. Even if the machine is in the same room, a serial console has huge advantages. Cut and paste between the consoles of seperate systems is incredibly useful.
The next step for the software giants would be threatening to sue discussion boards unless they remove negative posts about software. Microsoft could sue /. if someone posted benchmarks for SQLServer. Currently, it's unenforcable (though most of these sites lack the legal resources to oppose a big company). This law would make it enforcable. That would seriously damage our ability to pre-evaluate systems.
They could also try to sue Linus if a driver showed up that they felt was the result of reverse engineering their windows drivers.
When I started doing helpdesk and sysadmin work, users were expected to read the manuals. We would always answer their questions, but we also included a polite reference to the appropriate documentation. After a few examples, we only got calls about the obscure stuff.
I know that sounds harsh, but in the long run it's a kindness to both the user and your tech staff. The users understand things better, we have time to work at improving the system, and everyone's stress level is reduced.
Here's a paraphrase of an old proverb. If you give a man a fish, he eats for a day, then comes back tomorrow to mooch. If you teach him how to fish, he eats for a lifetime.
When did users decide they didn't need to read the docs, not even the brief field descriptions on the screen? More importantly, when did tech support groups decide to coddle stupid lusers with one time answers, instead of teaching them?
Traditionally free software projects have been paid for with peer respect and reputation. Open source generally preserves this. The problem has always been fuguring out how to use peer respect for grocerys. This system has the potential to solve that.
A project sponsor would probably consider the reputation of the bidding developers, and provide qualified ones with sample hardware. For example, almost any ethernet adaptor vendor would probably be willing to give a new card to Donald Becker.
The RIAA's next step would be to lobby for regulations that would require future mp3 players to detect the watermarks, and refuse to play the file. If they couldn't force it with legislation, they would subsidize manufacture of such devices. Their crippled players would be cheaper than the unlimited ones for long enough to drive the unlimited ones off the market.