I'm frequently asked to look at and evaluate various pieces of software. Usually, this is in preparation for organization-wide deployment and support.
However, <em>that's not my job</em>. It's a tiny little part of it. The server's broken, a user needs to have his machine fixed, and I need to evaluate your software. Guess which one goes to the bottom of my priority list.
When I actually have time to do evaluations, I work on two levels at once. I'm busy reading trade magazines, web sites, and usenet to <em>find</em> products, and as soon as I find one, I download it. I usually get an hour or two at a time to do this before something else grabs my attention, so unless your key arrives within 5 minutes, I've found another product to download and test. I don't get back to the evals for days at a time. If your registration process made me wait for more than a few minutes, I figure that I can wait a while to test the product, so it could be a few weeks before I finish gathering all of your competitors' products to testing and deploying yours.
If one of your fucking salesweasels has the nerve to call me the next day, you go to the bottom of the list. I'll happily tell the weasels that by making me wait for a registration key, they made me put off testing your product for a few weeks. If you call again, I mention that we include "harassing sales calls" in the "Service and Support" part of the evaluation, and every sales call reduces that company's score.
Test time. I have to dig through my email to find the key. Great. It's a 30 day key, and I downloaded your stuff 25 days ago. It'd better wow me in the next hour, since I probably won't have time for another look before the key expires. If I've heard so much as a peep from your sales people (i.e. the usual every-other-day "Didja try it yet? Do you like it? Huh? Huh?" bit), I'm not going to call and ask for a new key. I'm going to give it a zero for functionality ("Can't test it. Bad key.")and give your company a zero for service and support (for annoying me) and encourage my boss to go with a less annoying competitor.
Want to impress me? Do the following:
-Don't make me wait for a key
-If you must make me wait, give me a reasonably long time to test. Go ahead and lock the product to my domain or a small number of users if you want. Use the wait time to give me a custom demo copy.
-In your registration form, provide a "have one of your sales associates give me a call" checkbox, and <em>honor it</em>.
-Send follow-up literature by mail, and <em>make it useful</em>. I have half an hour to read it (well, I'll go back and read all of it if you're one of our top choices, but until then...). Glossy pictures of somebody else's server room and more than one buzzword per sentence won't impress me. Specs, limitations, and pricing will. Understand that in some organizations, the managers actually listen to the techies.
-Ask when we plan on deploying. If the answer is 6+ months, don't call me next week asking what I thought of your product. Call me in 3 months and ask if I had any problems.
You're confusing patents with trademarks. A trademark must be defended or it will be lost. You pay a fee to the patent office for an exclusive right to use your patent until it expires. You don't have to defend it unless you want to. The patent is still yours.
At the few places I've worked, the policy's always been that TTL = expected worst-case response time from the networking group plus a fudge factor.
So, if DNS goes down at 10:00pm on a Friday, people (who have the addresses cached) can still get to the machines until the hung-over networking crew logs in to check things out the next morning.
They'd bump the TTL way down, on the other hand, when major machine moves were planned.
A nice finished wood exterior, big-ass speaker in the main enclosure, and a wall projector in place of the dial. You'll put this giant enclosure next to the easy chair in the living room. It'll communicate with your other entertainment equipment via a sophisticated but quasi-proprietary wireless interface. It will be the home entertainment device that everyone has been dreaming of. And it'll cost a fortune.
Meanwhile, Microsoft will come out with a cheap plastic box that you can stick between your DVD player and your receiver. It'll have many of the same features, but it'll look worse, crash more, and cost 20% of what the Apple version costs.
Everybody will want the Apple version, but when it comes time to buy, they'll come home with the Microsoft version.
You're right. Mea culpa. It's not posession of stolen property. It's outright illegal according to a different set of laws: US Code Title 17.
USC Title 17, Section 106 says what exclusinve rights a copyright holder has. One of them is the right "to reproduce the copyrighted work in copies or phonorecords."
It's legal to download ROMs and keep them for as long as you want, mp3s or any other copyrighted content as well.
Uh, no.
Last I checked, "posession of stolen property" was still a crime.
It's PROBABLY legal to download ROMs and mp3s IF you also own the original copyrighted material. And even then, it's a little bit shaky.
You're allowed to make backups. You can copy CDs to tape or mp3 in case the original is damaged. You could argue that downloading the mp3s from audiogalaxy was the same as ripping the CD yourself. In fact, with the new anti-ripping stuff that new CDs are using, this claim becomes easier to make. However, it's still unclear enough that you might be charged and have to make your argument for a judge.
This was, in turn, based on "Shockwave", by Alvin Toffler.
It's the first mention that I've seen of computer worms. He assumed that people would use their telephones to access centralized and interconnected mainframe systems, and that both legitimate and malicious users would unleash worms through the systems to do their bidding. It also has a really interesting subplot about universal ID numbers.
The article claims that the disks hold 500MB, or enough for 11 CD-quality albums.
Uh, right.
A CD-quality album is 600MB. Compressing it down to fit in 45MB results in casette-tape quality music: good enough for portable devices or computer speakers, but lousy for a good stereo system. For that, you need significantly more storage space.
These shows aren't meant to show features in their final form. They're more "here's the nifty concept that's inspiring us right now."
So, some -useful- things that might come out of this:
When the car detects anger, it cuts acceleration and top speed, giving the driver fewer chances to drive like an asshole.
The same thing could (and probably should) be done right now when it detects the driver using a cellphone.
Combined with a big mp3 drive, you can just hit the "music" button without taking your eyes off the road to fiddle with the radio to find the right music for your current mood.
You could adjust traction control on 4WD vehicles to the driver's stress level. If they're an inch of snow on the ground, the Chicago driver won't even care, but the Dallas driver is going to have a death-grip on the wheel and'll be on the verge of panic. For the latter, engage the 4WD automatically, cut power so that they're less likely to spin, and put the anti-lock brakes on "paranoid mode."
One of the problems with collision detection systems is that they're really annoying during rush hour, and there actually are situations where driving close to the vehicle in front of you. You could link these systems together so that if the driver seems alert, the system won't go off, or will go off quietly, but if the driver's dealing with fighting kids in the back seat, it sets off the "too close" klaxon at full volume.
Re:In case you didn't notice their home page...
on
Loki Goes Postal
·
· Score: 2
Tribes 2 is done and out
Yes, it is. Don't get it if you're using a 3dfx card unless you enjoy updating significant chunks of your video software from CVS. At least, that was the situation when it came out.
Tribes 2 was enlightening for me. I enjoy using Linux. In fact, I provide Linux support as part of my job (and I still enjoy Linux;). I spent weeks getting it working properly. Aadmittedly, I was only spending a little bit of time on it each day, but I knew what I was doing and it took a while. For a user who's just taken the Linux plunge, perhaps using a burned copy of Red Hat 6.2 (7.x is still suckful, and a new user's likely to grab an old version from the bargain bin anyway), I sure as hell wouldn't want to talk them through the steps they'd have to follow to run the game. In fact, before I had it working under Linux, I'd given up and purchased the Windows version to play until I could un-kink the Linux version.
It was the first time in a while that I stopped and said "This is why people aren't buying linux games."
I think it's good that Loki's staying away from 3d acceleration right now. It's the latest trend in games, but 3d under Linux is still growing too fast for "normal users" to keep up.
Actually, the PSC started as a joint venture between The University of Pittsburgh, Carnegie-Mellon University, and Westinghouse. (Trivia: their BITNET nodes were all prefixed with CPWSC for Carnegie Pitt Westinghouse Supercomputing Center: CPWSCA, CPWSCB, etc.) Practically, this meant that Westinghouse supplied the machine room space, CMU provided the staff space and handled administrative stuff (like paychecks) and, well, Pitt students and professors could work at PSC and deal with Pitt's payroll system instead of CMU's.
The PSCs networking group spun off into the National Center for Network Engineering (NCNE) when PSC proper lost it's NSF funding (due to sheer administrative stupidity, IMO). The PSC provides the internet connection to CMU, Pitt, and Penn State, but they have no other real affiliation with Penn State.
They're pretty liberal about giving out research accounts, so long as you're somehow affiliated with one of their funding sources. Currently, this mostly means that you have to be an academic of some sort in Pennsylvania (including students) or you have to be doing networking research, particularly for Internet 2.
IIS has much much more functionality than Apache does, and it has been around much longer, unfortunately in this case longer means a more convoluted codebase,
Please name one bit of functionality that IIS had that apache does not. The only thing I can think of is.asp, and that's because Microsoft wanted a proprietary way to do the things that Apache users were already doing with perl and php.
The second bit is just insane. IIS was microsoft's late entry into the webserver wars, long after Apache was created. Apache, in turn, was "a patch-y" version of the old NCSA web server. I was going to get dates, but the NCSA httpd web pages haven't been updated since '96. There's some history here, though. The IIS code base is convoluted mostly because they were rushing to catch up so that people didn't give money to Netscape for their Windows-based web servers.
Personally, I have two primary email addresses: one for work, and one for personal stuff.
It amazes me that a lot of office workers don't consider this option. It's like only having a work-provided telephone, or a single mailing address where the letter from your aunt is mixed in with the latest HR newsletter.
When I leave the office on Friday, my work email stays there. There are escalation procedures if they need to contact me in a true emergency, but I don't respond to the minor problems. When I return to work, I check my email 2-3 times a day. If you respond to the inbox bell with pablovian conditioning, you won't get anything done. I read email, decide on the most important thing to do next, then do it. I don't check email until it's done or I'm at a good stopping point. Yes, there are the panicky nitwits who call if I don't respond in 5 minutes. It only takes a few rounds of "Is this really so important that it can't wait an hour?" followed by "I just read the message, and it CAN wait an hour. Click." before they get the point.
I treat personal email the same way. My friends know my phone number, and they know that I might not check or respond to email immediately. It confuses some of them, but they cope. They understand that I have two addresses, and if they send me something at 10:00am, I'm not going to read it until 6:00pm or whenever I'm not at work any more.
You just need to learn to break the cycle. I spend all day on the computer. I used to be a slave to my email. It was burning me out, so I stopped. The transition will piss a few people off, but in the end, you'll be happier and more productive if you don't check your email every few minutes.
how we do it
on
Dorm Storm?
·
· Score: 5, Informative
At the university where I work, we've been gearing up for the last few weeks. We have guides that answer the common questions for the users intelligent enough to read them. For the rest, we'll have every warm body helping with phones or going from room to room to help with setup.
One of the most important bits: have a clear SLA. Be sure that you know and users know exactly what you do and don't support. At this point, inconsistency is a killer, because if one guy's willing to do more than the others, users will keep calling back until they get that one guy. If anything's changed since last spring, be sure that <em>everyone</em> knows exactly what was changed and why.
Give your specialists some cross training. Be sure that your mac guys can do basic windows troubleshooting, and vice versa. It seems like all the Mac questions hit at once. It must be a mac user group mind thing.;)
It's too late for this year, but automate as much as you can for next year. If you give your users access to your help database and you give them documentation, a few will check there. Set up web forms for network registration, account registration, etc.
Whenever your department doesn't do something, find out who does, and make sure that your info's correct. Students will call IT wanting to know how to register for classes online, or how to set up their telephone. That might be enrollment or the registrar or telecom or someone else. Be sure that you know, and that it's documented so that you're not sending users on wild goose chases. Otherwise, they'll call back (or worse, be referred back by another clueless department), and the second time around, they'll be pissed.
Most importantly, schedule breaks. We tend to push ourselves too hard during this time of the year. A lot of people just keep going "for another five minutes" until they pass out because they've been working for 6 hours straight without stopping for food or toilet breaks. If you've got someone who won't stop, force them to get coffee for everyone else. That'll get them away from the users for a minute, at least.
Great. This'll make it even harder to find the cURL page in search engines.
cURL is a library that transfers files via protocols that use URL syntax. Given the trademark happy nature of the web site mentioned in this article, I suspect that there'll be a name fight in the future.
I've been thinking about this myself. Basically, we have a system where someone has to be responsible for each node on the internet. By providing public access, you are responsible for the traffic that flows through your node. So, you want to minimize harmful traffic. I see two ways to do this:
1) Proxied free access to port 80 outbound only. This way, you get rid of the spammers and slow down the script kiddies. Eventually, someone will end up using it for fraud of some sort, so it'd be good to use a proxy to at least prove that it wasn't you. Your ISP will boot you, but good logs might keep you out of jail.
2) Quasi-free access. A setup like the above that redirects any HTML request to a web page that asks people to sign up. Use the standard list management routine of "give me an email address. I'll send mail with a code. You return that code to me." Link that to a hardware address. Then, give registered hardware addresses access to anywhere. Again, through the proxy to save your butt if they do something bad. That way, you can at least hand the police an email address.
I think that subscriptions, not micropayments, are the way to go. Consider Slashdot. On an average day, I see a story or two that look really interesting, and they are interesting. I see a few more that might be interesting, but they have misleading headlines, they're not what I thought they were, etc. And, there are a few that just don't interest me at all. If I had to pay per-page or per-click, slashdot wouldn't be worth it. There's too much crap between the gems. However, over the course of a week/month/year, I find a lot of useful info, and the site's worthwhile. I see this same pattern repeated across every site I visit.
The problem with most web subscriptions is that they're overpriced. The web was supposed to bring good content at a low price because there was no middleman or shipping. However, many of the subscription sites are "content" sites (Salon, WSJ, Economist) who want to charge almost as much as I'd pay for a paper magazine. And, let's face it, the dead tree version is a whole lot more convenient. If I could get Salon for $10/year, I'd sign up in a heartbeat. I'd pay the same for slashdot. However, for $30+ year, I think more carefully about what I'm getting, and I usually decide against a subscription.
Suppose that the university offered a class on reverse engineering. Now, suppose that as part of this course, your next-to-final project involved writing code that couldn't be easily reverse engineered, and that the final project involved reverse engineering the binaries of your fellow students. If a student submitted their works under the GPL, the professor could incur legal liability for not also handing out the source, thus defeating the point of the entire exercise.
That example had a lot of perhapses and supposes, but consider something simpler: A student writes a simple product tracking database that suits the university's hardware auditing needs perfectly. The university deploys it, changes and enhances it, and then loses the source code through some sort of screwup. Failed disk, bad backups, etc. (Yes, it does happen.) Now, the university is distributing binaries of GPL'd code without making the source available. Legally, they have to yank software that's still useful. This seems to fall afoul of the agreement with the university.
This article ignores one of the findamental differences between software and other types of products: it can be copied at no cost. When I give you a copy of free software that I've written, I lost nothing. Nobody takes anything from me, and parasites do not damage or weaken the host in any way. These arguments are often repeated by software pirates, but that doesn't make them any less true.
When I write and release software, trust doesn't enter into it. It's my gift to the world. Eric Raymond's comparison of free software to other gift economies is very accurate for me. Take what I've made and use it. Make the world a better place. If it has to be proprietary, so be it.
This sort of unconditional gift isn't possible with the GPL, so I use the BSD license. As long as there are a few others doing the same, we can keep it up forever. This isn't a competition. We can all win.
Slashdot Mirror
I'm frequently asked to look at and evaluate various pieces of software. Usually, this is in preparation for organization-wide deployment and support.
However, <em>that's not my job</em>. It's a tiny little part of it. The server's broken, a user needs to have his machine fixed, and I need to evaluate your software. Guess which one goes to the bottom of my priority list.
When I actually have time to do evaluations, I work on two levels at once. I'm busy reading trade magazines, web sites, and usenet to <em>find</em> products, and as soon as I find one, I download it. I usually get an hour or two at a time to do this before something else grabs my attention, so unless your key arrives within 5 minutes, I've found another product to download and test. I don't get back to the evals for days at a time. If your registration process made me wait for more than a few minutes, I figure that I can wait a while to test the product, so it could be a few weeks before I finish gathering all of your competitors' products to testing and deploying yours.
If one of your fucking salesweasels has the nerve to call me the next day, you go to the bottom of the list. I'll happily tell the weasels that by making me wait for a registration key, they made me put off testing your product for a few weeks. If you call again, I mention that we include "harassing sales calls" in the "Service and Support" part of the evaluation, and every sales call reduces that company's score.
Test time. I have to dig through my email to find the key. Great. It's a 30 day key, and I downloaded your stuff 25 days ago. It'd better wow me in the next hour, since I probably won't have time for another look before the key expires. If I've heard so much as a peep from your sales people (i.e. the usual every-other-day "Didja try it yet? Do you like it? Huh? Huh?" bit), I'm not going to call and ask for a new key. I'm going to give it a zero for functionality ("Can't test it. Bad key.")and give your company a zero for service and support (for annoying me) and encourage my boss to go with a less annoying competitor.
Want to impress me? Do the following:
-Don't make me wait for a key
-If you must make me wait, give me a reasonably long time to test. Go ahead and lock the product to my domain or a small number of users if you want. Use the wait time to give me a custom demo copy.
-In your registration form, provide a "have one of your sales associates give me a call" checkbox, and <em>honor it</em>.
-Send follow-up literature by mail, and <em>make it useful</em>. I have half an hour to read it (well, I'll go back and read all of it if you're one of our top choices, but until then...). Glossy pictures of somebody else's server room and more than one buzzword per sentence won't impress me. Specs, limitations, and pricing will. Understand that in some organizations, the managers actually listen to the techies.
-Ask when we plan on deploying. If the answer is 6+ months, don't call me next week asking what I thought of your product. Call me in 3 months and ask if I had any problems.
Just make sure that you're not using ext2 and you either turn off cached writes or hook your NAS into an UPS.
NetBSD supports MOP and can be used to boot a DECstation (both the older MIPS based ones and the handful of Alpha based ones that use MOP).
You're confusing patents with trademarks. A trademark must be defended or it will be lost. You pay a fee to the patent office for an exclusive right to use your patent until it expires. You don't have to defend it unless you want to. The patent is still yours.
Wired magazine talked about this a while ago. The archived article is here
At the few places I've worked, the policy's always been that TTL = expected worst-case response time from the networking group plus a fudge factor.
So, if DNS goes down at 10:00pm on a Friday, people (who have the addresses cached) can still get to the machines until the hung-over networking crew logs in to check things out the next morning.
They'd bump the TTL way down, on the other hand, when major machine moves were planned.
Obviously, it'll look like an old radio
A nice finished wood exterior, big-ass speaker in the main enclosure, and a wall projector in place of the dial. You'll put this giant enclosure next to the easy chair in the living room. It'll communicate with your other entertainment equipment via a sophisticated but quasi-proprietary wireless interface. It will be the home entertainment device that everyone has been dreaming of. And it'll cost a fortune.
Meanwhile, Microsoft will come out with a cheap plastic box that you can stick between your DVD player and your receiver. It'll have many of the same features, but it'll look worse, crash more, and cost 20% of what the Apple version costs.
Everybody will want the Apple version, but when it comes time to buy, they'll come home with the Microsoft version.
USC Title 17, Section 106 says what exclusinve rights a copyright holder has. One of them is the right "to reproduce the copyrighted work in copies or phonorecords."
The definition of "copies" is given in and very clearly includes making a copy of a file on a hard drive. Basically, anything that you can listen to again later is a copy. An mp3 qualifies.
Penalties are given in USC Title 17, Chapter 5
Uh, no.
Last I checked, "posession of stolen property" was still a crime.
It's PROBABLY legal to download ROMs and mp3s IF you also own the original copyrighted material. And even then, it's a little bit shaky.
You're allowed to make backups. You can copy CDs to tape or mp3 in case the original is damaged. You could argue that downloading the mp3s from audiogalaxy was the same as ripping the CD yourself. In fact, with the new anti-ripping stuff that new CDs are using, this claim becomes easier to make. However, it's still unclear enough that you might be charged and have to make your argument for a judge.
Problem is, you're increasing the search engine scores when people search for KMPG.
Instead, we should be making them show up at the top of the list for more interesting things like goatfuckers, child porn, or just corporate fraud.
This was, in turn, based on "Shockwave", by Alvin Toffler.
It's the first mention that I've seen of computer worms. He assumed that people would use their telephones to access centralized and interconnected mainframe systems, and that both legitimate and malicious users would unleash worms through the systems to do their bidding. It also has a really interesting subplot about universal ID numbers.
The article claims that the disks hold 500MB, or enough for 11 CD-quality albums.
Uh, right.
A CD-quality album is 600MB. Compressing it down to fit in 45MB results in casette-tape quality music: good enough for portable devices or computer speakers, but lousy for a good stereo system. For that, you need significantly more storage space.
These shows aren't meant to show features in their final form. They're more "here's the nifty concept that's inspiring us right now."
So, some -useful- things that might come out of this:
When the car detects anger, it cuts acceleration and top speed, giving the driver fewer chances to drive like an asshole.
The same thing could (and probably should) be done right now when it detects the driver using a cellphone.
Combined with a big mp3 drive, you can just hit the "music" button without taking your eyes off the road to fiddle with the radio to find the right music for your current mood.
You could adjust traction control on 4WD vehicles to the driver's stress level. If they're an inch of snow on the ground, the Chicago driver won't even care, but the Dallas driver is going to have a death-grip on the wheel and'll be on the verge of panic. For the latter, engage the 4WD automatically, cut power so that they're less likely to spin, and put the anti-lock brakes on "paranoid mode."
One of the problems with collision detection systems is that they're really annoying during rush hour, and there actually are situations where driving close to the vehicle in front of you. You could link these systems together so that if the driver seems alert, the system won't go off, or will go off quietly, but if the driver's dealing with fighting kids in the back seat, it sets off the "too close" klaxon at full volume.
Yes, it is. Don't get it if you're using a 3dfx card unless you enjoy updating significant chunks of your video software from CVS. At least, that was the situation when it came out.
Tribes 2 was enlightening for me. I enjoy using Linux. In fact, I provide Linux support as part of my job (and I still enjoy Linux ;). I spent weeks getting it working properly. Aadmittedly, I was only spending a little bit of time on it each day, but I knew what I was doing and it took a while. For a user who's just taken the Linux plunge, perhaps using a burned copy of Red Hat 6.2 (7.x is still suckful, and a new user's likely to grab an old version from the bargain bin anyway), I sure as hell wouldn't want to talk them through the steps they'd have to follow to run the game. In fact, before I had it working under Linux, I'd given up and purchased the Windows version to play until I could un-kink the Linux version.
It was the first time in a while that I stopped and said "This is why people aren't buying linux games."
I think it's good that Loki's staying away from 3d acceleration right now. It's the latest trend in games, but 3d under Linux is still growing too fast for "normal users" to keep up.
Actually, the PSC started as a joint venture between The University of Pittsburgh, Carnegie-Mellon University, and Westinghouse. (Trivia: their BITNET nodes were all prefixed with CPWSC for Carnegie Pitt Westinghouse Supercomputing Center: CPWSCA, CPWSCB, etc.) Practically, this meant that Westinghouse supplied the machine room space, CMU provided the staff space and handled administrative stuff (like paychecks) and, well, Pitt students and professors could work at PSC and deal with Pitt's payroll system instead of CMU's.
The PSCs networking group spun off into the National Center for Network Engineering (NCNE) when PSC proper lost it's NSF funding (due to sheer administrative stupidity, IMO). The PSC provides the internet connection to CMU, Pitt, and Penn State, but they have no other real affiliation with Penn State.
They're pretty liberal about giving out research accounts, so long as you're somehow affiliated with one of their funding sources. Currently, this mostly means that you have to be an academic of some sort in Pennsylvania (including students) or you have to be doing networking research, particularly for Internet 2.
Please name one bit of functionality that IIS had that apache does not. The only thing I can think of is .asp, and that's because Microsoft wanted a proprietary way to do the things that Apache users were already doing with perl and php.
The second bit is just insane. IIS was microsoft's late entry into the webserver wars, long after Apache was created. Apache, in turn, was "a patch-y" version of the old NCSA web server. I was going to get dates, but the NCSA httpd web pages haven't been updated since '96. There's some history here, though. The IIS code base is convoluted mostly because they were rushing to catch up so that people didn't give money to Netscape for their Windows-based web servers.
I guess I have to add something to make the lameness filter happy. First time I've ever hit it. How annoying.
It amazes me that a lot of office workers don't consider this option. It's like only having a work-provided telephone, or a single mailing address where the letter from your aunt is mixed in with the latest HR newsletter.
When I leave the office on Friday, my work email stays there. There are escalation procedures if they need to contact me in a true emergency, but I don't respond to the minor problems. When I return to work, I check my email 2-3 times a day. If you respond to the inbox bell with pablovian conditioning, you won't get anything done. I read email, decide on the most important thing to do next, then do it. I don't check email until it's done or I'm at a good stopping point. Yes, there are the panicky nitwits who call if I don't respond in 5 minutes. It only takes a few rounds of "Is this really so important that it can't wait an hour?" followed by "I just read the message, and it CAN wait an hour. Click." before they get the point.
I treat personal email the same way. My friends know my phone number, and they know that I might not check or respond to email immediately. It confuses some of them, but they cope. They understand that I have two addresses, and if they send me something at 10:00am, I'm not going to read it until 6:00pm or whenever I'm not at work any more.
You just need to learn to break the cycle. I spend all day on the computer. I used to be a slave to my email. It was burning me out, so I stopped. The transition will piss a few people off, but in the end, you'll be happier and more productive if you don't check your email every few minutes.
At the university where I work, we've been gearing up for the last few weeks. We have guides that answer the common questions for the users intelligent enough to read them. For the rest, we'll have every warm body helping with phones or going from room to room to help with setup.
;)
One of the most important bits: have a clear SLA. Be sure that you know and users know exactly what you do and don't support. At this point, inconsistency is a killer, because if one guy's willing to do more than the others, users will keep calling back until they get that one guy. If anything's changed since last spring, be sure that <em>everyone</em> knows exactly what was changed and why.
Give your specialists some cross training. Be sure that your mac guys can do basic windows troubleshooting, and vice versa. It seems like all the Mac questions hit at once. It must be a mac user group mind thing.
It's too late for this year, but automate as much as you can for next year. If you give your users access to your help database and you give them documentation, a few will check there. Set up web forms for network registration, account registration, etc.
Whenever your department doesn't do something, find out who does, and make sure that your info's correct. Students will call IT wanting to know how to register for classes online, or how to set up their telephone. That might be enrollment or the registrar or telecom or someone else. Be sure that you know, and that it's documented so that you're not sending users on wild goose chases. Otherwise, they'll call back (or worse, be referred back by another clueless department), and the second time around, they'll be pissed.
Most importantly, schedule breaks. We tend to push ourselves too hard during this time of the year. A lot of people just keep going "for another five minutes" until they pass out because they've been working for 6 hours straight without stopping for food or toilet breaks. If you've got someone who won't stop, force them to get coffee for everyone else. That'll get them away from the users for a minute, at least.
cURL is a library that transfers files via protocols that use URL syntax. Given the trademark happy nature of the web site mentioned in this article, I suspect that there'll be a name fight in the future.
1) Proxied free access to port 80 outbound only. This way, you get rid of the spammers and slow down the script kiddies. Eventually, someone will end up using it for fraud of some sort, so it'd be good to use a proxy to at least prove that it wasn't you. Your ISP will boot you, but good logs might keep you out of jail.
2) Quasi-free access. A setup like the above that redirects any HTML request to a web page that asks people to sign up. Use the standard list management routine of "give me an email address. I'll send mail with a code. You return that code to me." Link that to a hardware address. Then, give registered hardware addresses access to anywhere. Again, through the proxy to save your butt if they do something bad. That way, you can at least hand the police an email address.
--
The problem with most web subscriptions is that they're overpriced. The web was supposed to bring good content at a low price because there was no middleman or shipping. However, many of the subscription sites are "content" sites (Salon, WSJ, Economist) who want to charge almost as much as I'd pay for a paper magazine. And, let's face it, the dead tree version is a whole lot more convenient. If I could get Salon for $10/year, I'd sign up in a heartbeat. I'd pay the same for slashdot. However, for $30+ year, I think more carefully about what I'm getting, and I usually decide against a subscription.
--
Yuck! /usr/local is for LOCAL software. Not managed ports stuff. I'd like to see something like /usr/ports for that stuff.
--
Suppose that the university offered a class on reverse engineering. Now, suppose that as part of this course, your next-to-final project involved writing code that couldn't be easily reverse engineered, and that the final project involved reverse engineering the binaries of your fellow students. If a student submitted their works under the GPL, the professor could incur legal liability for not also handing out the source, thus defeating the point of the entire exercise.
That example had a lot of perhapses and supposes, but consider something simpler: A student writes a simple product tracking database that suits the university's hardware auditing needs perfectly. The university deploys it, changes and enhances it, and then loses the source code through some sort of screwup. Failed disk, bad backups, etc. (Yes, it does happen.) Now, the university is distributing binaries of GPL'd code without making the source available. Legally, they have to yank software that's still useful. This seems to fall afoul of the agreement with the university.
--
When I write and release software, trust doesn't enter into it. It's my gift to the world. Eric Raymond's comparison of free software to other gift economies is very accurate for me. Take what I've made and use it. Make the world a better place. If it has to be proprietary, so be it.
This sort of unconditional gift isn't possible with the GPL, so I use the BSD license. As long as there are a few others doing the same, we can keep it up forever. This isn't a competition. We can all win.
--