Disable WEP and don't worry about it. WEP is easily cracked and, more importantly, it encourages bad habits - if you use strong end to end encryption, you don't need it; if you don't use encryption, your data is being sent in plain-text from the point it leaves your WAP.
We're keeping old HP printers around because they work. Sure, they're slow, the postscript support is flaky and the network stack is garbage but you can work around those. It's much more of a hassle to use the newer HPs which have jams and other mechanical failures on a regular basis. We've gone through all of the usual procedures, had them professionally serviced, etc. - they're just poorly designed.
Unfortunately, there's not much connection between cost and quality - expensive workgroup laser printers seem to jam about as often as cheap deskjets. HP's firmware hasn't improved much, either - the newer printers don't hang if they get multiple simultaneous connections but they still go into/dev/null mode and choke on some postscript documents - and they continue to be quite slow - I've never seen anything close to the rated speed in actual usage since the processors aren't even remotely capable of keeping up with the print engine once you get past the "hello world" level. PDFs containing complex figures are measured in minutes per page even on the "workgroup" printers.
There are two new printers I rely on: a very expensive Canon ImageRunner copier which doubles as the uber-printer and a Xerox / Tektronix Phaser 8200, which is a color wax printer. Both have been rock-solid, handled all sorts of convoluted jobs and are *much* faster than the latest HPs - the ImageRunner is rated at 60 pages per minute and I've never seen much less, even with huge files containing truly vile postscript. This isn't surprising - it has an 800MHz PIII instead of the slow 300Mhz ARM/MIPS-class CPU which is all HP can afford to put in a $16,000 printer.
Unseemly, opportunistic, even dishonorable but it looks like it was technically legal. That's why I get annoyed when people start ranting about the "stolen" election - it may be fun to vent but it distracts attention away from the underlying problems and makes it harder to get mainstream support. ("Voting machine improvements? That's just another whiny democrat who can't accept that he lost!")
Exactly - it's *scary* how many "rights" depend on the continued good will of an elected body. The only thing which could hold Parliament back would be one of the EU agreements they signed - that's the reason for the official state of emergency after 9/11 since it freed them of the obligation to follow it...
That's not true - the system worked about as well as you could expect given that the difference was well under its margin of error. Gore won with some techniques and lost with others - the answer isn't to sit around whining as if the will of some vast majority of the people was thwarted but to work to eliminate such embarassingly bad polling techniques.
The bigger problem is that most people did not vote - democracy works only as well as the voters. Right now life in the US is still comfortable enough for people to get away with abrogating their responsibilities but that can't continue forever.
Scalia is not insurmountable. The real part will be getting the pro-civil liberties portion of the Democratic support base to work with the small-government portion of the Republican party. I think a concerted campaign could carry elections but I'm dubious about it coming together.
There have been a number of legal reviews which have concluded that the Europeans are keeping pace with the US on that front. The situation is actually worse in like England where any right can be revoked by the current government - at least in the US you at least have the hope of getting something truly egregious thrown out as unconstitutional.
Speaking of which, it's probably time to start planning for some protracted legal battles cleaning up the anti-terrorist mess.
Like most/. types, I'm frequently pulling down files from all sorts of different servers. I'll pick HTTP any time I have the choice as my experience has been that FTP is far more likely to be slow and unreliable.
I'd prefer to admin Apache rather than any FTP server I've used, simply because it's more flexible.
There's also a technical problem with FTP - the split control / data feature. Firewalls rarely break HTTP; FTP is far more likely to get complaints from users. Most of the servers are also rather low quality, too - small HTTP transfers almost always complete before an FTP server finishes connecting, let alone transfering a file, and I've run into a surprising number which don't allow you to resume transfers.
All the time - even four years ago when I was relying on HTTP's resume to work around a cable modem which would hang connections after 50KB transfered (don't ask), it didn't require anything unusual.
I think it would be more accurate to say that the 2D APIs and hardware haven't advanced at anything like the pace 3D acceleration has enjoyed. Windows doesn't use the 3D hardware to do compositing and there are a number of operations you can get for free with modern hardware.
There are quite a few people interested in using the 3D hardware for 2D tasks. See this post for a comparison between GDI and D3D for 2D work.
Back in my COBOL days I took some time getting used to just how conservative large financial institutions are. Even the most conservative mainstream software companies are raging cowboys in comparison.
Even if you port and test everything, they're going to wait until there's a substantial track record of working reliably simply because no ever wants to find an obscure condition which incorrectly bills a million people - even a minor rounding error can be significant with billions of dollars floating around. They're going to do anything necessary to avoid having to prove the fault-tolerance system any of the thousands of transactions in a momentary outage from being dropped or (worse) misprocessed.
The other factor is constraints - there are a surprising number of contracts, regulations, industry rules, etc. which spell out the exact environment something is going to run on. Getting changes approved can take absurd amounts of time. The change management process on this kind of large system will seem completely unreal.
I get a fair number of these messages (for some reason my yahoo.com address gets joe-jobbed every couple weeks) and found procmail / spamassassin extremely useful. Simple From/To filtering isn't reliable any more but content based filtering has a lot of life left in it - SpamAssassin will block most bounces based on the included spam in common bounce formats and procmail or perl's Mail::Audit are sufficiently flexible to get whatever's left if there's anyway to meaningfully filter them, as is usually the case.
The next step is time-based addresses - perhaps having the Evil Bastard filter on your generic foo@example.com and having a bypass for key@example.com, where the key rotates every few days. Finally, you could have your filter drop bounces which don't contain an email address and subject matching email in your sent mail folder or use a custom keyed return-path and drop bounces which don't use it.
Unfortunately, most of this is impractial for people who don't run their own mail servers. SpamAssassin is at least available as a plugin for anyone stuck with Outlook, so there's hope that more advanced filtering will sneak in to common use.
One other note - if you can figure out who was responsible file a case in small claims court. They'll lose by default if they don't show up (which is almost certain if they aren't local) and you give a default judgement to a collection agency for a percentage of the award.
Cheap to some people means a couple hundred dollars
Check eBay - people have been getting 100Mb switches w/fiber uplinks for as little as $10 recently. One big switch for the main office and a couple little ones for the other end would be doable for $200.
I'm not sure about the USB modems but the PC Card modems work with anything which supports PCMCIA serial devices and PPP - it appears as a stock dialup connection using 3333 as the phone number with no username or password.
Because Ricochet is faster (1xRTT's real-world 40-60Kbs is a at best third of the speed), cheaper ($45 / mo vs. $99, no need for a new phone) and easier to connect your PC to. In my experience, Ricochet takes a minute to setup on a new laptop, whereas I know people who've had to spend hours with working through the various permutations of settings and updates to get 1xRTT working (this will eventually improve as the software and hardware mature - it's a classic 1.0 release).
It's quite easy and a very good idea from a security perspective. Unfortunately, it can be amazingly unreliable if you aren't really careful with your hardware/os combinations.
If it changes *every* variable there's no difference between "Tom" and "VAR000002" - neither variable will have its original name in the end. This also illustrates why this problem isn't just a repeated string replacement.
This would be an excellent addition for a package management system - when you install foo.(deb|rpm) it could automatically put a set of sane defaults in some master directory under/etc which could be extended (or overriden if the sysadmin allows it) by a file in a similar directory (.sandbox?) in your home directory.
It really gets down to time management: would those programmers take the time saved by using a less-primitive language and use it to fix the design problems behind those slow programs or add a bunch of pointless crap?
I think the latter is more likely, no matter how much we might wish otherwise. Almost all of the slow programs we use are slow because of dumb code (from simple implementation gaffs up to complete architectural clusterfucks) and that's a language independent "skill".
Slashdot Mirror
Disable WEP and don't worry about it. WEP is easily cracked and, more importantly, it encourages bad habits - if you use strong end to end encryption, you don't need it; if you don't use encryption, your data is being sent in plain-text from the point it leaves your WAP.
We're keeping old HP printers around because they work. Sure, they're slow, the postscript support is flaky and the network stack is garbage but you can work around those. It's much more of a hassle to use the newer HPs which have jams and other mechanical failures on a regular basis. We've gone through all of the usual procedures, had them professionally serviced, etc. - they're just poorly designed.
/dev/null mode and choke on some postscript documents - and they continue to be quite slow - I've never seen anything close to the rated speed in actual usage since the processors aren't even remotely capable of keeping up with the print engine once you get past the "hello world" level. PDFs containing complex figures are measured in minutes per page even on the "workgroup" printers.
Unfortunately, there's not much connection between cost and quality - expensive workgroup laser printers seem to jam about as often as cheap deskjets. HP's firmware hasn't improved much, either - the newer printers don't hang if they get multiple simultaneous connections but they still go into
There are two new printers I rely on: a very expensive Canon ImageRunner copier which doubles as the uber-printer and a Xerox / Tektronix Phaser 8200, which is a color wax printer. Both have been rock-solid, handled all sorts of convoluted jobs and are *much* faster than the latest HPs - the ImageRunner is rated at 60 pages per minute and I've never seen much less, even with huge files containing truly vile postscript. This isn't surprising - it has an 800MHz PIII instead of the slow 300Mhz ARM/MIPS-class CPU which is all HP can afford to put in a $16,000 printer.
Still, that's only half the speed of what I'm seeing from download.rhn.redhat.com and you didn't have to pay for the privilege of a slow download.
Unseemly, opportunistic, even dishonorable but it looks like it was technically legal. That's why I get annoyed when people start ranting about the "stolen" election - it may be fun to vent but it distracts attention away from the underlying problems and makes it harder to get mainstream support. ("Voting machine improvements? That's just another whiny democrat who can't accept that he lost!")
Exactly - it's *scary* how many "rights" depend on the continued good will of an elected body. The only thing which could hold Parliament back would be one of the EU agreements they signed - that's the reason for the official state of emergency after 9/11 since it freed them of the obligation to follow it...
That's not true - the system worked about as well as you could expect given that the difference was well under its margin of error. Gore won with some techniques and lost with others - the answer isn't to sit around whining as if the will of some vast majority of the people was thwarted but to work to eliminate such embarassingly bad polling techniques.
The bigger problem is that most people did not vote - democracy works only as well as the voters. Right now life in the US is still comfortable enough for people to get away with abrogating their responsibilities but that can't continue forever.
Scalia is not insurmountable. The real part will be getting the pro-civil liberties portion of the Democratic support base to work with the small-government portion of the Republican party. I think a concerted campaign could carry elections but I'm dubious about it coming together.
There have been a number of legal reviews which have concluded that the Europeans are keeping pace with the US on that front. The situation is actually worse in like England where any right can be revoked by the current government - at least in the US you at least have the hope of getting something truly egregious thrown out as unconstitutional.
Speaking of which, it's probably time to start planning for some protracted legal battles cleaning up the anti-terrorist mess.
Like most /. types, I'm frequently pulling down files from all sorts of different servers. I'll pick HTTP any time I have the choice as my experience has been that FTP is far more likely to be slow and unreliable.
I'd prefer to admin Apache rather than any FTP server I've used, simply because it's more flexible.
There's also a technical problem with FTP - the split control / data feature. Firewalls rarely break HTTP; FTP is far more likely to get complaints from users. Most of the servers are also rather low quality, too - small HTTP transfers almost always complete before an FTP server finishes connecting, let alone transfering a file, and I've run into a surprising number which don't allow you to resume transfers.
You're correct except for point 1. All of the major HTTP servers allow you to limit simultaneous connections or (better) bandwidth.
All the time - even four years ago when I was relying on HTTP's resume to work around a cable modem which would hang connections after 50KB transfered (don't ask), it didn't require anything unusual.
I think it would be more accurate to say that the 2D APIs and hardware haven't advanced at anything like the pace 3D acceleration has enjoyed. Windows doesn't use the 3D hardware to do compositing and there are a number of operations you can get for free with modern hardware.
There are quite a few people interested in using the 3D hardware for 2D tasks. See this post for a comparison between GDI and D3D for 2D work.
Back in my COBOL days I took some time getting used to just how conservative large financial institutions are. Even the most conservative mainstream software companies are raging cowboys in comparison.
Even if you port and test everything, they're going to wait until there's a substantial track record of working reliably simply because no ever wants to find an obscure condition which incorrectly bills a million people - even a minor rounding error can be significant with billions of dollars floating around. They're going to do anything necessary to avoid having to prove the fault-tolerance system any of the thousands of transactions in a momentary outage from being dropped or (worse) misprocessed.
The other factor is constraints - there are a surprising number of contracts, regulations, industry rules, etc. which spell out the exact environment something is going to run on. Getting changes approved can take absurd amounts of time. The change management process on this kind of large system will seem completely unreal.
I get a fair number of these messages (for some reason my yahoo.com address gets joe-jobbed every couple weeks) and found procmail / spamassassin extremely useful. Simple From/To filtering isn't reliable any more but content based filtering has a lot of life left in it - SpamAssassin will block most bounces based on the included spam in common bounce formats and procmail or perl's Mail::Audit are sufficiently flexible to get whatever's left if there's anyway to meaningfully filter them, as is usually the case.
The next step is time-based addresses - perhaps having the Evil Bastard filter on your generic foo@example.com and having a bypass for key@example.com, where the key rotates every few days. Finally, you could have your filter drop bounces which don't contain an email address and subject matching email in your sent mail folder or use a custom keyed return-path and drop bounces which don't use it.
Unfortunately, most of this is impractial for people who don't run their own mail servers. SpamAssassin is at least available as a plugin for anyone stuck with Outlook, so there's hope that more advanced filtering will sneak in to common use.
One other note - if you can figure out who was responsible file a case in small claims court. They'll lose by default if they don't show up (which is almost certain if they aren't local) and you give a default judgement to a collection agency for a percentage of the award.
Check eBay - people have been getting 100Mb switches w/fiber uplinks for as little as $10 recently. One big switch for the main office and a couple little ones for the other end would be doable for $200.
I'm not sure about the USB modems but the PC Card modems work with anything which supports PCMCIA serial devices and PPP - it appears as a stock dialup connection using 3333 as the phone number with no username or password.
Because Ricochet is faster (1xRTT's real-world 40-60Kbs is a at best third of the speed), cheaper ($45 / mo vs. $99, no need for a new phone) and easier to connect your PC to. In my experience, Ricochet takes a minute to setup on a new laptop, whereas I know people who've had to spend hours with working through the various permutations of settings and updates to get 1xRTT working (this will eventually improve as the software and hardware mature - it's a classic 1.0 release).
In San Diego, I see 180Kbs routinely now; with Metricom I used to see 220Kbs and often hit 300Kbs+, a level I have yet to see with the new service.
Please give the court a copy of the police report
It's quite easy and a very good idea from a security perspective. Unfortunately, it can be amazingly unreliable if you aren't really careful with your hardware/os combinations.
If it changes *every* variable there's no difference between "Tom" and "VAR000002" - neither variable will have its original name in the end. This also illustrates why this problem isn't just a repeated string replacement.
The people who actually make the engines don't seem to agree with you:
turbo diesel intercooler: 0 hits
Compare to turbocharged direct injection
Laid-off divxnetworks employee, right?
This would be an excellent addition for a package management system - when you install foo.(deb|rpm) it could automatically put a set of sane defaults in some master directory under /etc which could be extended (or overriden if the sysadmin allows it) by a file in a similar directory (.sandbox?) in your home directory.
It really gets down to time management: would those programmers take the time saved by using a less-primitive language and use it to fix the design problems behind those slow programs or add a bunch of pointless crap?
I think the latter is more likely, no matter how much we might wish otherwise. Almost all of the slow programs we use are slow because of dumb code (from simple implementation gaffs up to complete architectural clusterfucks) and that's a language independent "skill".