Others have said I'm sure, but buying a broadband router with included firewall (without wireless) can be had for $20-40.
Plug one of those beasts into the wall between the machine and the network, and you can be almost guaranteed that the machines are going to survive the latest attacks.
When you notice someone is infected, drop them off the network.
When they complain, tell them that they need to fix their computer before they can get online again. Say they can acquire the tools online for free from any internet enabled computer, or they can pay the $5 and you can make them a bootable linux CD + F-Prot Linux (do it once, upgrade as necessary).
After they have run the software, they call you, you turn them on.
If they are infected again, you drop them again.
Alternatively, you can shutdown the university network, and allow DSL and cablemodem companies to service the users. If they want professional service, they can pay a professional; people in the real world do.
I called up tech support when I was using all three of them, and they suggested doing that same thing. Unfortunately (for me and them), it didn't help the situation. One had all lights always on, one had all lights always flashing, one had all lights always off. Maybe I have bad luck with batteries.
I bought the 2nd generation from a Best Buy last summer, and it didn't work (all lights were always on, but wouldn't put any power into my laptop, an IBM R51). After exchanging it, a similar problem reared its ugly head. I exchanged it a second time, and the third battery didn't work either. I returned it and decided that I was done with it until they made a 3rd generation.
Pick up a cheap old laptop with a couple PCMCIA (aka PC-Card) slots. Grab a pair of old 3COM network adapters (with the dongles so you can run one incoming and one outgoing), and you have a low-power router. A P75 or high-end 486 would be enough to route and do mail for you.
The hardware would set you back no more than around $100-200.
As long as you're not compiling on the little guy, he'll do just fine - I know more than a few people who have run linux and BSDs on 486s as NAT routers for 3mbit cablemodems.
I'm not sure if anyone else has posted something similar previous to this, but on Myth Busters (a show on the Discovery channel) they stuck two dead pigs in a Corvette for like 2 weeks.
Result: no amount of cleaning, commercial solvents, enzymes, etc., was able to remove the smell. They ended up selling the thing for scrap for around $1000. The new owner was going to pull the engine, tires, and various other components. All metal and plastic compents that had been exposed to the stench retained the smell.
I'd personally buy all new, transfer the data, wipe the drives, and find some way of disposing of the equipment; the smell isn't going to go away.
What you are basically saying is that you offer various packages for various prices, and if someone doesn't like the packages, they can go somewhere else. Regardless of copyrights, it is all about terms of the contract.
I agree with you. Give them options. If they don't like options, or think they can get the same or better service for less, they can go somewhere else.
Goodness, you cited my editor (PyPE). I don't know if I would call it mature, but "works well enough" and "pretty much bug free" are two ways that I would describe it. Oh, that and "I've been using it daily since it was 6 hours old" is another.
Ahh, such is the power of Python development. You can have a usable application (source code editor) in a weekend.
I once had the job of doing basically the same thing as the asker was asked to do, though I was to scan 176 pages in total.
I ended up getting a login on an old dual PPro 200, running NT4, with an old HP SCSI scanner. Scanned in at 300 dpi, 256 color greyscale. Each page took around 10-15 seconds, each save and page swap took 15-20 seconds (I used the kodakimg app that came with NT, saving to compressed tiff). I initially tried the same thing with a USB scanner, but each page scan took 1-2 minutes. To hell with that.
After the 176 pages (88 pages double-sided), my arms and back were sore, but it wasn't too bad. Thankfully, I got them in two sets (115 and 61 pages), and it was even easier.
I did a scripted resize with photoshop to fit each image on its own page (though you could probably do the same thing with the 'resize image' powertoy for XP), generated a single web page that contained all of the images, loaded the web page (it took a few minutes), and used 'print to pdf' which is available if you have Acrobat installed.
If the asker only has one class, and its only 1-200 pages/week, 2-3 hours to do it all isn't that bad. No OCR, but I've seen so many crappy conversions it is hard for me to trust them.
In other forums I've stated this before, which still makes sense. I don't know if other people are taking this stance or not, so I'm going to go ahead and post it.
While I don't know the legal aspects of the license that was given to
the site via "Licence # LS-3M-02-36 of the Russian Multimedia and
Internet Society", most countries that license such things limit the
licensees by a stipulation that states, "you are only licensed to
distribute to people currently residing in this country", 'this country'
being the country of issue, and 'residing' meaning that they purchaser
only needs to have a physical presence. Generally, having a license to
sell something, doesn't necessarily mean that you have a license to
export.
If it does include such a clause, then distributing abroad (from Russia
to anywhere else) is in fact illegal, and at least in America,
"ignorance is no excuse for breaking the law", copyright or otherwise.
Generally if one really had no idea, there is a lenient penalty (except
in the case when the RIAA comes knocking, or when the crime was horrendous). However, if one was conscious
of the questionable legality of something, that is pretty much an admission of guilt.
If it doesn't include a clause, then I suppose this email was basically
pointless. Though I will point out (which you do as well) that for the
price that they are selling tracks, I doubt much, if any of the proceeds
goes to the artists themselves. This moves us from a legal conundrum to
a moral one.
I would imagine that the industry would feel that distributing digital copies of games, even if it were more secure than 1024 bit RSA encryption, would be cutting its own nuts off. I'd be willing to bet a few dollars that this is one of the many reasons why the Phantom game console isn't going to make it.
On the longevity side, current burnable CDs have lifetimes of ~5 years under decent conditions, where as pressed CDs, as long as they aren't scratched, have lifetimes in the 30+ range. While we can't guarantee that our OSes will last that long, it would be nice if the media our games were on lasted.
Pick up a copy of stunnel for both machines, set up an ssl tunnel for the control port (21), and let the data channel be unencrypted. Then you don't have to worry about entering a password twice (SSH tunnels require once for ssh and once for ftp).
I think it just works (in a similar fashion to the way ssh tunnels for ftp just work). But as I've not tried it, I don't know for certain.
What gets me is that this guy thinks this book deserves a 10. Apparently he really didn't read Snow Crash, because it was actually awesome.
Certainly I have not read Decipher, but from his plot summary I can tell you that it sounds like crap. Checking the first Editorial Review here: http://www.amazon.com/exec/obidos/tg/detail/-/0312 280750//qid=1059161726/sr=2-1/ref=sr_2_1/002-37184 57-9016815?v=glance&s=books&vi=reviews really tells me that it IS crap.
The reviewer comparing this to Neal Stephenson should be ashamed of himself.
Yeah...file size limits are the result of drive formatting, not how many bits of addressing the machine can handle. I've got a 32 bit machine, yet I have a 6 gig file (growing to 13.6 over the course of the next few hours). That is well beyond your '2 gig limit'. Most decent operating systems and programming languages running on the system will return a 64 bit integer for file sizes, because most modern operating systems (with a modern filesystem) support file sizes larger than 2 gigs.
There's an article that was linked from slashdot describing how to convert access to sql here: http://developers.slashdot.org/developers/03/04/14/1059254.shtml?tid=185&tid=156
About a year and a half ago, The Offspring released Original Prankster on mp3.com saying that if it could hit top 30 on MTV, they would release another track for free. If it got up higher, another...etc.
The longer it stayed on the top MTV charts, the more potential songs released (there may have been a mention of full-cd release if #1 for some amount of time).
Check it out, they've got 3 tracks up for download. http://artists.mp3s.com/artists/166/the_offspring. html
When I first found C/C++ in my freshman year of college, I was in love. It just made sense with the way my mind worked at the time, and I could just hammer away and make good working code quickly.
A year later I found Python. All those things you want, quick development time, limited bugs, etc., I have found to be accomplished through the use of Python. I've written high-speed socket servers, GUI applications (with wxPython), console email readers... So many things that all work well, and I spent more time writing them than debugging them. And writing them took 1/10th the time of C/C++ development. That coupled with amazing built-in and user developed modules...really makes Python my language of choice when doing ANY development, including processor-intensive stuff.
As long as I program in the proper Python style, I find that my application development time is reduced significantly, the incidence of bugs are very low (less than 1 per 10 lines of code, usually syntax errors - forgot a paren or colon), and maintainability is high.
Popfile was announced here in late August, shortly after the Paul Graham article came out. It was originally closed source, which prompted the creation of multiple other projects. Among them is Spambayes and even my own Pasp (both in python, both open source).
As well, Popfile was announced open source at the end of September...on Slashdot. I know this because it was released under such a license as I was finishing up Pasp.
So yeah. As for how well Popfile categorizes mail into multiple categories, I have not run many tests with multiple category bayesian filtering, though the Spambayes group has, and has discovered that filtering mail based on multiple categories is far less accurate (many false categorizations). In the minimal tests I have done, I find this to be the case as well (we are used to less than 2% FP and FN rates, and with >2 bin categorization, error rates spike easily into the 10% range).
So yeah. Popfile has been announced here no less than 3 times now. I've not seen Spambayes announced at all (they deserve it), and Pasp has also not been announced, though I could care less about that.
I wrote a paper in a similar vein last spring about stopping ddos attacks, it's the second section of this paper. It seeks to fix the underlying problem, not create a band-aid.
Slashdot Mirror
Agreed, install Postfix, and hire yourself a competant Postfix admin for 10 hours/week (whether he works them or not).
Others have said I'm sure, but buying a broadband router with included firewall (without wireless) can be had for $20-40.
Plug one of those beasts into the wall between the machine and the network, and you can be almost guaranteed that the machines are going to survive the latest attacks.
That's good to know. Though I don't think my wife will give up her standard wired optical mouse, even for the Wacom mouse.
Wacom pads have been doing this thing for years, for both the pens and the mouse (though they use absolute positioning).
When you notice someone is infected, drop them off the network.
When they complain, tell them that they need to fix their computer before they can get online again. Say they can acquire the tools online for free from any internet enabled computer, or they can pay the $5 and you can make them a bootable linux CD + F-Prot Linux (do it once, upgrade as necessary).
After they have run the software, they call you, you turn them on.
If they are infected again, you drop them again.
Alternatively, you can shutdown the university network, and allow DSL and cablemodem companies to service the users. If they want professional service, they can pay a professional; people in the real world do.
That too.
I called up tech support when I was using all three of them, and they suggested doing that same thing. Unfortunately (for me and them), it didn't help the situation. One had all lights always on, one had all lights always flashing, one had all lights always off. Maybe I have bad luck with batteries.
I bought the 2nd generation from a Best Buy last summer, and it didn't work (all lights were always on, but wouldn't put any power into my laptop, an IBM R51). After exchanging it, a similar problem reared its ugly head. I exchanged it a second time, and the third battery didn't work either. I returned it and decided that I was done with it until they made a 3rd generation.
Pick up a cheap old laptop with a couple PCMCIA (aka PC-Card) slots. Grab a pair of old 3COM network adapters (with the dongles so you can run one incoming and one outgoing), and you have a low-power router. A P75 or high-end 486 would be enough to route and do mail for you.
The hardware would set you back no more than around $100-200.
As long as you're not compiling on the little guy, he'll do just fine - I know more than a few people who have run linux and BSDs on 486s as NAT routers for 3mbit cablemodems.
I'm not sure if anyone else has posted something similar previous to this, but on Myth Busters (a show on the Discovery channel) they stuck two dead pigs in a Corvette for like 2 weeks.
Result: no amount of cleaning, commercial solvents, enzymes, etc., was able to remove the smell. They ended up selling the thing for scrap for around $1000. The new owner was going to pull the engine, tires, and various other components. All metal and plastic compents that had been exposed to the stench retained the smell.
I'd personally buy all new, transfer the data, wipe the drives, and find some way of disposing of the equipment; the smell isn't going to go away.
What you are basically saying is that you offer various packages for various prices, and if someone doesn't like the packages, they can go somewhere else. Regardless of copyrights, it is all about terms of the contract.
I agree with you. Give them options. If they don't like options, or think they can get the same or better service for less, they can go somewhere else.
Goodness, you cited my editor (PyPE). I don't know if I would call it mature, but "works well enough" and "pretty much bug free" are two ways that I would describe it. Oh, that and "I've been using it daily since it was 6 hours old" is another.
Ahh, such is the power of Python development. You can have a usable application (source code editor) in a weekend.
Thank you for pimping it.
Point for you, but a SCSI scanner is still good advice.
I once had the job of doing basically the same thing as the asker was asked to do, though I was to scan 176 pages in total.
I ended up getting a login on an old dual PPro 200, running NT4, with an old HP SCSI scanner. Scanned in at 300 dpi, 256 color greyscale. Each page took around 10-15 seconds, each save and page swap took 15-20 seconds (I used the kodakimg app that came with NT, saving to compressed tiff). I initially tried the same thing with a USB scanner, but each page scan took 1-2 minutes. To hell with that.
After the 176 pages (88 pages double-sided), my arms and back were sore, but it wasn't too bad. Thankfully, I got them in two sets (115 and 61 pages), and it was even easier.
I did a scripted resize with photoshop to fit each image on its own page (though you could probably do the same thing with the 'resize image' powertoy for XP), generated a single web page that contained all of the images, loaded the web page (it took a few minutes), and used 'print to pdf' which is available if you have Acrobat installed.
If the asker only has one class, and its only 1-200 pages/week, 2-3 hours to do it all isn't that bad. No OCR, but I've seen so many crappy conversions it is hard for me to trust them.
In other forums I've stated this before, which still makes sense. I don't know if other people are taking this stance or not, so I'm going to go ahead and post it.
While I don't know the legal aspects of the license that was given to the site via "Licence # LS-3M-02-36 of the Russian Multimedia and Internet Society", most countries that license such things limit the licensees by a stipulation that states, "you are only licensed to distribute to people currently residing in this country", 'this country' being the country of issue, and 'residing' meaning that they purchaser only needs to have a physical presence. Generally, having a license to sell something, doesn't necessarily mean that you have a license to export.
If it does include such a clause, then distributing abroad (from Russia to anywhere else) is in fact illegal, and at least in America, "ignorance is no excuse for breaking the law", copyright or otherwise. Generally if one really had no idea, there is a lenient penalty (except in the case when the RIAA comes knocking, or when the crime was horrendous). However, if one was conscious of the questionable legality of something, that is pretty much an admission of guilt.
If it doesn't include a clause, then I suppose this email was basically pointless. Though I will point out (which you do as well) that for the price that they are selling tracks, I doubt much, if any of the proceeds goes to the artists themselves. This moves us from a legal conundrum to a moral one.
I would imagine that the industry would feel that distributing digital copies of games, even if it were more secure than 1024 bit RSA encryption, would be cutting its own nuts off. I'd be willing to bet a few dollars that this is one of the many reasons why the Phantom game console isn't going to make it.
On the longevity side, current burnable CDs have lifetimes of ~5 years under decent conditions, where as pressed CDs, as long as they aren't scratched, have lifetimes in the 30+ range. While we can't guarantee that our OSes will last that long, it would be nice if the media our games were on lasted.
Pick up a copy of stunnel for both machines, set up an ssl tunnel for the control port (21), and let the data channel be unencrypted. Then you don't have to worry about entering a password twice (SSH tunnels require once for ssh and once for ftp).
I think it just works (in a similar fashion to the way ssh tunnels for ftp just work). But as I've not tried it, I don't know for certain.
What gets me is that this guy thinks this book deserves a 10. Apparently he really didn't read Snow Crash, because it was actually awesome.
2 280750//qid=1059161726/sr=2-1/ref=sr_2_1/002-37184 57-9016815?v=glance&s=books&vi=reviews
Certainly I have not read Decipher, but from his plot summary I can tell you that it sounds like crap. Checking the first Editorial Review here: http://www.amazon.com/exec/obidos/tg/detail/-/031
really tells me that it IS crap.
The reviewer comparing this to Neal Stephenson should be ashamed of himself.
Yeah...file size limits are the result of drive formatting, not how many bits of addressing the machine can handle. I've got a 32 bit machine, yet I have a 6 gig file (growing to 13.6 over the course of the next few hours). That is well beyond your '2 gig limit'. Most decent operating systems and programming languages running on the system will return a 64 bit integer for file sizes, because most modern operating systems (with a modern filesystem) support file sizes larger than 2 gigs.
But otherwise, you are correct.
There's an article that was linked from slashdot describing how to convert access to sql here: http://developers.slashdot.org/developers/03/04/14 /1059254.shtml?tid=185&tid=156
You know, it all depends.
. html
About a year and a half ago, The Offspring released Original Prankster on mp3.com saying that if it could hit top 30 on MTV, they would release another track for free. If it got up higher, another...etc.
The longer it stayed on the top MTV charts, the more potential songs released (there may have been a mention of full-cd release if #1 for some amount of time).
Check it out, they've got 3 tracks up for download. http://artists.mp3s.com/artists/166/the_offspring
Who else read that as "Porno comic"? Given the context, it is even funnier..."I like it." Yeah, you and all the other hentais out there.
FYI: Hentai is not so much a TYPE of anime erotica, so much as it describes those who watch it, 'Pervert'.
When I first found C/C++ in my freshman year of college, I was in love. It just made sense with the way my mind worked at the time, and I could just hammer away and make good working code quickly.
A year later I found Python. All those things you want, quick development time, limited bugs, etc., I have found to be accomplished through the use of Python. I've written high-speed socket servers, GUI applications (with wxPython), console email readers... So many things that all work well, and I spent more time writing them than debugging them. And writing them took 1/10th the time of C/C++ development. That coupled with amazing built-in and user developed modules...really makes Python my language of choice when doing ANY development, including processor-intensive stuff.
As long as I program in the proper Python style, I find that my application development time is reduced significantly, the incidence of bugs are very low (less than 1 per 10 lines of code, usually syntax errors - forgot a paren or colon), and maintainability is high.
I hate to mention this, but I will anyways.
Popfile was announced here in late August, shortly after the Paul Graham article came out. It was originally closed source, which prompted the creation of multiple other projects. Among them is Spambayes and even my own Pasp (both in python, both open source).
As well, Popfile was announced open source at the end of September...on Slashdot. I know this because it was released under such a license as I was finishing up Pasp.
So yeah. As for how well Popfile categorizes mail into multiple categories, I have not run many tests with multiple category bayesian filtering, though the Spambayes group has, and has discovered that filtering mail based on multiple categories is far less accurate (many false categorizations). In the minimal tests I have done, I find this to be the case as well (we are used to less than 2% FP and FN rates, and with >2 bin categorization, error rates spike easily into the 10% range).
So yeah. Popfile has been announced here no less than 3 times now. I've not seen Spambayes announced at all (they deserve it), and Pasp has also not been announced, though I could care less about that.
This sounds like Steve Gibson's suggestion from gibson research.
I wrote a paper in a similar vein last spring about stopping ddos attacks, it's the second section of this paper. It seeks to fix the underlying problem, not create a band-aid.