"Designing form-based web pages now requires beta testing against IE6, Netscape 7, and every version of Mozilla that ever existed."
There are some solutions which allow you to control the environment your scripts are running in... If you've got anything remotely suited to server-side, then it can remove many PITA (notably, it can save you from having to use Internet Explorer)
Plus, everyone who uses Internet Explorer needs to disable active scripting anyway, just to make it safe to browse the web with. So javascript stuff just stops working, until you get a browser that can deal with it safely, and give you the confidence to re-enable scripting with some limits.
"Turns out the problem was a combination of:Sun's JVM and my ATI video driver (which is a forcefit as Compaq never put out an XP driver for the model laptop I have)."
Was going to suggest checking Java if something is causing a Windows machine to die.
Of course, having just installed a soundcard on my machine at work and watched it bluescreen to the point where it wouldn't even boot in safe mode, and the system-repair CD itself finished off with a BSOD, my opinion of PC/Windows hardware stability is lessening...
"Who is the genius putting Windows on these things??"
The same genius using an Access database on the same machine without a password to store the votes.
From the diebold memos, the programmer doesn't seem too bright either, certainly has an unawareness of most security issues: "I could password-protect the database, but it would take ages to code, and I'd have to re-write the software to store the password"
Naturally, no mention of public-key systems, nor of translucent databases, has ever crossed the email-boxes of these people.
"Just tell me your votes" said the man behind the curtain, "I promise to tell you the right answer at the end"
so many "clearly-delineated" sections you can hardly find anything. It's 4 sections of which 2 are sponsored links, one is a search of their website, and one is a web-search.
Of course, the banner adverts on the top and bottom of the page are also clearly-delineated, just in case you wanted some colour to brighten up the day. What's the word for someone with a love of horizontal lines?
"RMS pushes free software as a moral imperative for all developers. Gates doesn't, so far as I have seen."
It's easy to look disinterested when the whole world is converted to your viewpoint. If Free Software was seen as the "normal" way of distributing software, then RMS wouldn't be doing much talking either.
Let's take a look back to when Free Software was normal, then:
In the early 1970s, Gates wrote the Open Letter to Hobbyists, which shocked the computer hobbyist community by insisting that a commercial market existed for computer software and that such software should not be freely copied without the publisher's permission. At the time, the community was strongly influenced by its ham radio legacy and the related Hacker ethic, which insist that innovations and knowledge should be freely shared in the community. Gates went on to co-found Microsoft Corporation, one of the world's most successful commercial enterprises, and led the way toward the emergence of the commercial software industry. -
wiki
"Has Gates ever actually come out and said that he thinks all software 'should' be proprietary?"
Name one piece of software he's written that isn't?
C'mon, head of the world's largest software company, thousands of different software products... if their owner wanted to release open software, you'd think somebody would have noticed it by now.
Oh look, a free wooden horse... We'd better bring that inside.
Ok look, a free Word-document reader. We'd better convert all our files to that format.
The only answer is to stop providing information, in case somebody reads it? The answer is to encourage ways of delivering advertising?
If people want to stop providing information, let them. Then their opinions will become irrelevant, outvoiced by the people who do want to provide information for the sake of it, rather than advertising dollars.
Subscription sites are the wet-dreams of advertising execs, not something that works in the real world, and certainly not something to use as a threat against people who read news but not advertising.
It doesn't seem to connect, you just get this gurgling tone from your modem, and in the background you can almost hear a little voice going "hello?..." or something.
"the act meets the definition of cyberterrorism since it endangered public safety"
Yep, we've said it all along: car drivers are terrorists.
Re:The Perfect Virus..?
on
The Virus Squad
·
· Score: 4, Informative
"I was thinking about how to design the "perfect" virus."
(1) Virus intially comes in as an attachment. This is a decoy, we're not going for computers owned by retards this time.
(2) Virus tests for one of the recent linux vulnerabilities. If it gets in, this indicates that we've got someone with a default unpatched install of Mandrake or whatever, who probably imagines they're immune. Plenty of time to proceed.
(3) Virus has a look through the setup files of common FTP programs to obtain website passwords, connects to website, searches for.exe and.tar.gz files, uploads itself in their place. Virus knows that people will download the.tar.gz, configure, make, and install it, then run it without even looking at the source code.
(4) Virus uploads a set of personal data to a hidden file on that website.
(5) Virus goes through the ~/Mail folder, looking for username/password combinations mailed to the person by clueless companies such as maplin.co.uk, who email peoples' passwords in cleartext. Stores a list of all the data it's collected so far.
(6) Virus sets up a backdoor, using port-knocking so that none of the "respond to virus with portscan" tools can find it.
"I'm guessing each of those 55 people doesn't make $220,000 a year."
It's what their time costs, not what you pay them. I was using the figure of GBP 10,000 per person per month, which should include their office-space, heating, lighting, computers, software, electricity, management, car-parking, insurance, healthcare, and all the other overhead costs it takes to keep 55 people working on a project.
Admittedly, multiplying that by the exchange rate to get $18000 is probably wrong, as it's not goods which are being traded, so I should probably start again with whatever the labour costs are in the US. But slashdot won't let me type the pound symbol, so I had to do something to convert it to a recognisable currency.
That number came from a management course in the UK, so again, correct it if it's wrong, but I don't think you can just multiply salary by time, unless you're talking about contractors working from home which we're not.
Based on permament employees working 40 hour weeks.
"If you think that the American government cares about the safety of foreign nationals, then you clearly haven't been paying attention to the news lately."
I'm not planning to visit America. To be honest, I'd feel safer in North Korea.
"But the city representatives know this, too, and I'm sure they are trying to exploit it to some degree"
Okay, before we all get excited, can I ask just one thing:
When was the last time you heard about a public-sector IT project that succeeded?
Linux or no, it seems like pretty much every IT investment made by any government (maybe it's just the UK?) ends up becoming a steaming pile of failure that gets reported ten-times overbudget, 8 years late, and eventually gets either scrapped or audited (those being essentially the same)
And it happens when they buy Windows software. All the time. Look at the UK's NHS information system. Read "computer weekly" or other rags. Nearly everything the government touches, fails.
So what happens when it happens again, and this time Linux was involved?
Will Microsoft publicise the news worldwide, and show that it was completely the fault of trying to use non-Microsoft software? Damn right. Will the public and the TV and the newspapers completely forget the hundreds of delayed, overbudget, and scrapped projects using Windows? Wouldn't put it past them.
It needs to be done anyway, but expecting a massive success seems rather optomistic. The best we can hope for is that hundreds of unpaid LUGs and volunteers end up making it work just to show off the hacker culture, and hopefully that'll be enough that the next person to upgrade their OS looks at it and thinks, "let's try that then..."
"What happens when there's a bicycle marathon on the cycle path during rush hour?:)"
If it's rush-hour, then the traffic is at a standstill anyway and it doesn't really matter whether the car-driver waits behind a red light, or drives through it and waits behind the traffic queue the other side.
They're going to take an hour to get through nottingham anyway if it's rush hour -- I'll probably go to the supermarket, do some shopping, and come back to find the same people waiting in the same queue.
Okay, I've read all that, and it pretty much confirms rather than allays my suspicions:
(a) the pedestrian sensor looks at the centre of the area where pedestrians would stand. But the buttons are right at the edge of that area. And cyclists can't move sideways.
(b) No mention of how long it takes from pressing a button to getting a green light. My guess is that it's "2 minutes, or when there's no traffic, whichever is earlier"
Which means that yes, the light does indeed fail to stop traffic if it can possibly avoid doing so. Remember that by the time the button is pushed, someone is already waiting
Having lights which take so long to react seems quite dangerous, because people will give up and cross anyway, and almost get hit by a police car
As to 'pre-timed max', surely imposing a "minimum time between stopping cars" doesn't have any effect if the crossing was just waiting there doing nothing for 10 minutes before I arrived? Unless it sets a minimum time for pedestrians to wait, which is just too dumb to even contemplate.
"We have migrated approximately 20000 users from 95 to 2000 in less than a year. Successfully."
Okay, somebody anonymously tell us how much that project cost then, so that we can compare the cost of a Windows upgrade to the cost of a Linux migration. It might be visible in your timesheet summaries, or you could ask whoever manages the timesheets.
Even estimating, you mention 55 people working for a year. I don't know if you work a U.S. 80-hour week, or a German 35-hour week, but that sounds suspiciously close to $12 million from where I'm sitting, or double that if you're paying someone else to do it.
That's manpower time of course, maybe $600 per user you've spent on support. So the manpower cost of an upgrade seems to be about equivalent to the cost of a Windows/Office license (ignoring any specialist tools that might need upgrading)
Of course, Windows upgraders will need that money to pay for licenses, while linux users will probably want a similar amount of money to get their favourite linux applications up to scratch. Naturally, this only needs to be done once if it's an open-source app -- and would it really take the full $10M to fix the remaining bugs in your favoute GNU/Linux application?
Challenge my figures if you've got better information of course, it's always interesting to find out how much these things really cost.
"So the button does nothing but you need to push it to cross legally."
Second-semester electronic engineering course at Nottingham University, the situation you describe was actually set out in a requirement when students were asked to program a traffic-light controller.
The module was using a Z180 to control a set of LED "traffic-lights", and I asked WTF would you want that sequence, it's inefficient and illogical.
Well I can't say it was the only illogical part of the course, but those requirements are certainly still being used.
Try Nottingham (the english one) for some really nice crossings, which actually detect a cyclist approaching on cycle-routes (embedded inductive sensor under the pavement) in enough time that it changes the lights in time for you to cross without even needing to slow down.
Now if only more junction designers could have a look at stuff like that, and see how convenient it is when things "just work"...
In England, we have these gutless pedestrian crossings which are too scared to stop traffic if they detect cars approaching, so they wait until there's no traffic around and only then activate the pedestrian sequence.
Well gee thanks, I could've figured out myself that I can get across when there's no cars around...
Even better are the ones with a sensor to see if a pedestrian is waiting. So not only do they pander to any approaching car, but they require the pedestrian to be standing in a particular place otherwise they don't operate at all. Very useful.
So if anyone is reading this story and doesn't have a clue what it means because traffic-light stuff is all greek to you... Bedford city council has jobs waiting for you. Join the ranks of the clueless.
"MS stuff intentionally doesn't play well with other software"
The obvious answer would be to start off with the right system, then whaddya know, no migration costs, you're already on linux.
With software though, it's always good to start early with an "our standard document format is sxw", which is easy enough without even having to mention the C-word [change]. Then allow a few months to get your documents in good shape. You know the saying, "information goes into DOC format, it doesn't come back out"
You can run lots of good software on Windows, even things like having POP/IMAP and SMTP mailservers instead of exchange ones, then pick a mail client you like. (unfortunately all the good mail clients are Linux only, but there're a few that work). Calendaring software and groupware is an easy one, and it gets easier if you don't have any legacy Exchange servers to support.
Just generally, make sure that "must natively handle microsoft file formats" is never a requirement for *any* purchase. Goodness knows how easy it would be to use your choice of software if people didn't put up "it doesn't do proprietry microsoft standard X" as an objection.
"I use wxWidgets (formerly wxWindows)"
How easy is it to get OpenGL working on that?
"Designing form-based web pages now requires beta testing against IE6, Netscape 7, and every version of Mozilla that ever existed."
There are some solutions which allow you to control the environment your scripts are running in... If you've got anything remotely suited to server-side, then it can remove many PITA (notably, it can save you from having to use Internet Explorer)
Plus, everyone who uses Internet Explorer needs to disable active scripting anyway, just to make it safe to browse the web with. So javascript stuff just stops working, until you get a browser that can deal with it safely, and give you the confidence to re-enable scripting with some limits.
"Turns out the problem was a combination of:Sun's JVM and my ATI video driver (which is a forcefit as Compaq never put out an XP driver for the model laptop I have)."
Was going to suggest checking Java if something is causing a Windows machine to die.
Of course, having just installed a soundcard on my machine at work and watched it bluescreen to the point where it wouldn't even boot in safe mode, and the system-repair CD itself finished off with a BSOD, my opinion of PC/Windows hardware stability is lessening...
"Who is the genius putting Windows on these things??"
The same genius using an Access database on the same machine without a password to store the votes.
From the diebold memos, the programmer doesn't seem too bright either, certainly has an unawareness of most security issues: "I could password-protect the database, but it would take ages to code, and I'd have to re-write the software to store the password"
Naturally, no mention of public-key systems, nor of translucent databases, has ever crossed the email-boxes of these people.
"Just tell me your votes" said the man behind the curtain, "I promise to tell you the right answer at the end"
"As long as the paid placements are delineated as such (e.g. Google's paid listings) they may not be such a bad thing."
Ever seen about.com?
so many "clearly-delineated" sections you can hardly find anything. It's 4 sections of which 2 are sponsored links, one is a search of their website, and one is a web-search.
Of course, the banner adverts on the top and bottom of the page are also clearly-delineated, just in case you wanted some colour to brighten up the day. What's the word for someone with a love of horizontal lines?
"Stephen Hawking and Kip Thorne may owe John Preskill a set of encyclopedias of his choice"
Do they take Wiki?
It's easy to look disinterested when the whole world is converted to your viewpoint. If Free Software was seen as the "normal" way of distributing software, then RMS wouldn't be doing much talking either.
Let's take a look back to when Free Software was normal, then:
"Except in writing poetry, if you throw out every rule in the book, you can create some masterpieces"
Sounds like Perl to me...
Of course, code isn't art if you have coding standards. It's more like plumbing or something.
"Has Gates ever actually come out and said that he thinks all software 'should' be proprietary?"
Name one piece of software he's written that isn't?
C'mon, head of the world's largest software company, thousands of different software products... if their owner wanted to release open software, you'd think somebody would have noticed it by now.
Oh look, a free wooden horse... We'd better bring that inside.
Ok look, a free Word-document reader. We'd better convert all our files to that format.
Sounds a bit tactical, no?
The only answer is to stop providing information, in case somebody reads it? The answer is to encourage ways of delivering advertising?
If people want to stop providing information, let them. Then their opinions will become irrelevant, outvoiced by the people who do want to provide information for the sake of it, rather than advertising dollars.
Subscription sites are the wet-dreams of advertising execs, not something that works in the real world, and certainly not something to use as a threat against people who read news but not advertising.
"unless 911 are offering PPP services now!"
It doesn't seem to connect, you just get this gurgling tone from your modem, and in the background you can almost hear a little voice going "hello?..." or something.
Dit.. dee.. nEEEEEEEEEEEEEEEEEEEEE... "sir?"
"the act meets the definition of cyberterrorism since it endangered public safety"
Yep, we've said it all along: car drivers are terrorists.
"I was thinking about how to design the "perfect" virus."
.exe and .tar.gz files, uploads itself in their place. Virus knows that people will download the .tar.gz, configure, make, and install it, then run it without even looking at the source code.
(1) Virus intially comes in as an attachment. This is a decoy, we're not going for computers owned by retards this time.
(2) Virus tests for one of the recent linux vulnerabilities. If it gets in, this indicates that we've got someone with a default unpatched install of Mandrake or whatever, who probably imagines they're immune. Plenty of time to proceed.
(3) Virus has a look through the setup files of common FTP programs to obtain website passwords, connects to website, searches for
(4) Virus uploads a set of personal data to a hidden file on that website.
(5) Virus goes through the ~/Mail folder, looking for username/password combinations mailed to the person by clueless companies such as maplin.co.uk, who email peoples' passwords in cleartext. Stores a list of all the data it's collected so far.
(6) Virus sets up a backdoor, using port-knocking so that none of the "respond to virus with portscan" tools can find it.
"I'm guessing each of those 55 people doesn't make $220,000 a year."
It's what their time costs, not what you pay them. I was using the figure of GBP 10,000 per person per month, which should include their office-space, heating, lighting, computers, software, electricity, management, car-parking, insurance, healthcare, and all the other overhead costs it takes to keep 55 people working on a project.
Admittedly, multiplying that by the exchange rate to get $18000 is probably wrong, as it's not goods which are being traded, so I should probably start again with whatever the labour costs are in the US. But slashdot won't let me type the pound symbol, so I had to do something to convert it to a recognisable currency.
That number came from a management course in the UK, so again, correct it if it's wrong, but I don't think you can just multiply salary by time, unless you're talking about contractors working from home which we're not.
Based on permament employees working 40 hour weeks.
"If you think that the American government cares about the safety of foreign nationals, then you clearly haven't been paying attention to the news lately."
I'm not planning to visit America. To be honest, I'd feel safer in North Korea.
"But the city representatives know this, too, and I'm sure they are trying to exploit it to some degree"
Okay, before we all get excited, can I ask just one thing:
When was the last time you heard about a public-sector IT project that succeeded?
Linux or no, it seems like pretty much every IT investment made by any government (maybe it's just the UK?) ends up becoming a steaming pile of failure that gets reported ten-times overbudget, 8 years late, and eventually gets either scrapped or audited (those being essentially the same)
And it happens when they buy Windows software. All the time. Look at the UK's NHS information system. Read "computer weekly" or other rags. Nearly everything the government touches, fails.
So what happens when it happens again, and this time Linux was involved?
Will Microsoft publicise the news worldwide, and show that it was completely the fault of trying to use non-Microsoft software? Damn right. Will the public and the TV and the newspapers completely forget the hundreds of delayed, overbudget, and scrapped projects using Windows? Wouldn't put it past them.
It needs to be done anyway, but expecting a massive success seems rather optomistic. The best we can hope for is that hundreds of unpaid LUGs and volunteers end up making it work just to show off the hacker culture, and hopefully that'll be enough that the next person to upgrade their OS looks at it and thinks, "let's try that then..."
"What happens when there's a bicycle marathon on the cycle path during rush hour? :)"
If it's rush-hour, then the traffic is at a standstill anyway and it doesn't really matter whether the car-driver waits behind a red light, or drives through it and waits behind the traffic queue the other side.
They're going to take an hour to get through nottingham anyway if it's rush hour -- I'll probably go to the supermarket, do some shopping, and come back to find the same people waiting in the same queue.
Okay, I've read all that, and it pretty much confirms rather than allays my suspicions:
(a) the pedestrian sensor looks at the centre of the area where pedestrians would stand. But the buttons are right at the edge of that area. And cyclists can't move sideways.
(b) No mention of how long it takes from pressing a button to getting a green light. My guess is that it's "2 minutes, or when there's no traffic, whichever is earlier"
Which means that yes, the light does indeed fail to stop traffic if it can possibly avoid doing so. Remember that by the time the button is pushed, someone is already waiting
Having lights which take so long to react seems quite dangerous, because people will give up and cross anyway, and almost get hit by a police car
As to 'pre-timed max', surely imposing a "minimum time between stopping cars" doesn't have any effect if the crossing was just waiting there doing nothing for 10 minutes before I arrived? Unless it sets a minimum time for pedestrians to wait, which is just too dumb to even contemplate.
"Seriously, these things ["look left" signs in London]are very useful when everyone is driving on the wrong side of the road."
So why don't they have them in countries where people drive on the wrong side of the road?
"We have migrated approximately 20000 users from 95 to 2000 in less than a year. Successfully."
Okay, somebody anonymously tell us how much that project cost then, so that we can compare the cost of a Windows upgrade to the cost of a Linux migration. It might be visible in your timesheet summaries, or you could ask whoever manages the timesheets.
Even estimating, you mention 55 people working for a year. I don't know if you work a U.S. 80-hour week, or a German 35-hour week, but that sounds suspiciously close to $12 million from where I'm sitting, or double that if you're paying someone else to do it.
That's manpower time of course, maybe $600 per user you've spent on support. So the manpower cost of an upgrade seems to be about equivalent to the cost of a Windows/Office license (ignoring any specialist tools that might need upgrading)
Of course, Windows upgraders will need that money to pay for licenses, while linux users will probably want a similar amount of money to get their favourite linux applications up to scratch. Naturally, this only needs to be done once if it's an open-source app -- and would it really take the full $10M to fix the remaining bugs in your favoute GNU/Linux application?
Challenge my figures if you've got better information of course, it's always interesting to find out how much these things really cost.
"So the button does nothing but you need to push it to cross legally."
Second-semester electronic engineering course at Nottingham University, the situation you describe was actually set out in a requirement when students were asked to program a traffic-light controller.
The module was using a Z180 to control a set of LED "traffic-lights", and I asked WTF would you want that sequence, it's inefficient and illogical.
Well I can't say it was the only illogical part of the course, but those requirements are certainly still being used.
Try Nottingham (the english one) for some really nice crossings, which actually detect a cyclist approaching on cycle-routes (embedded inductive sensor under the pavement) in enough time that it changes the lights in time for you to cross without even needing to slow down.
Now if only more junction designers could have a look at stuff like that, and see how convenient it is when things "just work"...
In England, we have these gutless pedestrian crossings which are too scared to stop traffic if they detect cars approaching, so they wait until there's no traffic around and only then activate the pedestrian sequence.
Well gee thanks, I could've figured out myself that I can get across when there's no cars around...
Even better are the ones with a sensor to see if a pedestrian is waiting. So not only do they pander to any approaching car, but they require the pedestrian to be standing in a particular place otherwise they don't operate at all. Very useful.
So if anyone is reading this story and doesn't have a clue what it means because traffic-light stuff is all greek to you... Bedford city council has jobs waiting for you. Join the ranks of the clueless.
"MS stuff intentionally doesn't play well with other software"
The obvious answer would be to start off with the right system, then whaddya know, no migration costs, you're already on linux.
With software though, it's always good to start early with an "our standard document format is sxw", which is easy enough without even having to mention the C-word [change]. Then allow a few months to get your documents in good shape. You know the saying, "information goes into DOC format, it doesn't come back out"
You can run lots of good software on Windows, even things like having POP/IMAP and SMTP mailservers instead of exchange ones, then pick a mail client you like. (unfortunately all the good mail clients are Linux only, but there're a few that work). Calendaring software and groupware is an easy one, and it gets easier if you don't have any legacy Exchange servers to support.
Just generally, make sure that "must natively handle microsoft file formats" is never a requirement for *any* purchase. Goodness knows how easy it would be to use your choice of software if people didn't put up "it doesn't do proprietry microsoft standard X" as an objection.
"What's the price of freedom, Mr. Ballmer?"
Apparently you can buy the "freedom to innovate" for $100,000 per year