I'm not sure what you're getting at with this question, but I don't think developing this system, or this kind of thinking in general, is bad.
We slam agencies (Government, FAA) when they are not proactive enough in protecting the general public, as we accuse them of implementing change only in response to bloodshed when a major tragedy occurs. Are we also going to complain when someone is forward-looking into protecting the public against something that has never happened (yet), but could?
The thought of a plane full of people being shot down by an RPG is not that wild and stupid. Do we need to wait for it to happen before looking into solutions? Of course it will cost billions. But can you put a price on your [daughter|son|mother|father|loved one] who could just be sitting on that plane?
I don't think it's a sense of distrust... But then again, when I got my license to drive at 17, my parents only very reluctantly lent me their car twice. Despite being on my best behaviour both times, they made it clear that if I wanted to drive, I would have to buy my own car.
So I got a part time job, borrowed money from Dad and bought a decent used car, and paid back every penny of my loan. And hey, the car was mine, so I was free to do as I please. Abuse it? Break it? Too bad, I had to fix it.
Despite that, I still managed to earn a college diploma and today, at 33, I have a wife, kids, 3 cars, a nice house that's paid for and a great job.
I think letting your kids drive around in a vehicle that doesn't belong to them reduces their incentive to work and pay for what they want. Has nothing to do with responsibility. You want it, earn it!
Actually, McDonalds doesn't claim to have served "people" - they simply claim "xx billions served". In this case, I think it's more logical to assume they're refering to orders.
As long as Mozilla do like McDonalds - claim "millions of downloads", not "millions of people downloaded" they'll be okay.
I use SATA on our smaller, non-mission-critical servers. For our data backend, I wouldn't touch it with a 10-foot pole.
Here are some scenarios where I wouldn't hesitate to use SATA:
- You have redundant servers. Using LVS and/or Heartbeat and your favorite tools, you can get full server redundancy using less expensive hardware. The overall solution can be quite elegant, with hot failover. Why just cover the drives?
- Front-end cluster nodes. You have a powerful, expensive backend server (with a cheaper failover) and you use inexpensive front-end servers for serving client requests. Sounds like overkill for what you want, but with the right server load balancing technology, it can give you a scalable, fault-tolerant and damn fast solution.
- You can live with downtime. Install a server with a couple of SATA disks in a RAID configuration and hope for the best.
It is your right to tell me to "screw off" if I suggest an improvement - after all, I am free to take my dollars somewhere else if I don't like your site (hereinafter "Product"). Having said that, I like your Product, but I'd like to offer my opinion:
my most-important-link rule I agree with you.
Next is proper anchor texting. I agree with you.
I should not be allowed to post my views. I consider this opinion to be simply ludicrous. I agree. Again, if people don't like it, your readership will go down. Right now your Product is wildly popular, so this is not an issue.
spelling and grammar. Let me be clear. As you are probably well aware, I don't think these are as important as the things I mentioned above. I disagree with you. Your Product consists of one-paragraph stories, and you only post 50 in a day - they deserve to be properly written and spell-checked. The way I see it, so many people are asking for better spelling, and it would result in a better Product. Are we all wrong? You are under no obligation to improve your Product, but in today's competitive world, if a better Product comes along, you may lose customers. That's up to you.
doesn't really matter much. People found other things to complain about. Yes, some folks are annoying. I tend to take it with a grain of salt: if one person complains about everything about your Product, then your Product is not for them. If everyone complains about the same thing, then ignoring those complaints is being short-sighted (or close-minded).
We get millions of e-mails a day, and everyone complains We don't care - quit your whining. Every job on this planet is hard. You are offering a Product, and if delivering said Product is too much work, you have many options: a) quit b) make processes more efficient c) hire more staff d) delegate: levrage the power of your community. Have the community edit articles. You can make it work. What? No money to modify Slashcode? Have call for contributions. A code camp to improve Slashcode. A fundraising campaign. I'd gladly contribute $10 towards making Slashdot a more community-driven site. Right now.
I've been coming here for eight years, and Taco, you *still* seem to take criticism too personally. I was accused of that not long ago. When something's your baby and it's popular, you tend to focus on the negative.
You can't please everybody so quit trying. Enjoy what you're doing for the millions who like it, who use it and who don't say a word. But...
Don't hesitate to listen to those who say something's broken. If one person sais it's broken, laugh - it's funny. When one hundred say it's broken, suck it up - it most likely is. BUT that doesn't mean it's gawd-awful and it's certainly NOT a mandate to fix it. If you don't have the resources to make it better, so be it. Don't beat yourself up. Add it to the FAQ.
Be more transparent. If we knew what your workday looked like, maybe we'd understand. As it is, we don't.
Be fair with yourself, and with your readers. Like above, many bitch and complain, but we're not always wrong. Likewise, we don't know the whole story behind Slashdot, so keep that in mind. When someone sais you should post reject notices to stories, they have no clue what that involves. Chuckle and add it to the FAQ. Life sucks for them.
A sample size of 1 is adequate when you only have one server - and you run the benchmark on the actual server you're going to be using.
Having a benchmark use a sample of 1000 different units and tell you that ReiserFS 4 is the best is useless if it performs worst on your specific hardware platform.
An interesting analysis in every aspect, and it's fine and dandy for the person who uses 400 GB drives and a ATA controller on a 500MHz computer but I'd like to see how the filesystems compare on a bigass RAID system run by a Power5 server, or a few Itaniums that usually have with a few hundred connected users. Something a bit more "entreprise" - where the choice of a filesystem is a bit more critical than a small server or a home PC.
Back in the day, I think the only reason Novell tacked on a GUI to NetWare was because of the pressure they were feeling from Windows NT and the new wave of "Admins" who were addicted to the mouse. NetWare 3.12/4.x was *the* file and print solution, and NDS is just so many miles ahead of Active Directory.
We run a cluster of SLES 9 servers, consisting of 4 Itanium2's, 4 x86_64's, 3 IBM Power5's and one x86, and they run flawlessly. No X, no KDE crap, just a barebones minimal install + the 25-or-so server packages we need. Keeping the OS up-to-date has been flawless as well, including Kernel upgrades.
It's nice to have a unified OS for all the hardware platforms we run.
Here's what I think he'll answer to your first question:
" MySQL works fine on Slashdot. It has all the features and performance we need, it has been running flawlessly for years and we're already familiar with it, so why should we change to anything else? What makes you think there's a bullet that needs biting? Granted, PG looks neat and all, but why exchange a dollar for four quarters? "
Here's what I think he'll answer to your second question:
" MySQL 5 doesn't offer us any features we absolutely need (otherwise we'd be using PG, right?) We will upgrade eventually, but we have bigger fish to fry right now, and upgrading our database is not very imperative. "
Something else he might say:
" Running slashdot is not as simple as running a basement website that gets 3 hits per hour. Thought needs to be put into these decisions. We can't just run off and install something the day it's released. "
ethtool and mii-tool. The Cisco 2970 switch we use has the knack of initializing the interface in half-duplex mode if the port is set to "Auto". Easily fixed by setting the port to Full, but useful nonetheless.
strings. Good to check if executables are using/etc/hosts.allow and/etc/hosts.deny, or where pop3/imap are looking for PEM certificates to configure pop3s/imaps, etc...
vmstat. Think your system is paging, or a card is generating too many interrupts?/proc filesystem. Favorites include/proc/net/dev, and/proc/uptime,/proc/cpuinfo,/proc/loadavg. Good for aggregating individual server load data in a cluster.
awk and sed. Mentioned elsewhere, but priceless.
chmod. I think the Linux filesystem permissions are too... permissive. Cut down on access to your logfiles and to config files in/etc.
*quota*. A must for restricting disk space use.
umask. When you need root, set your default umask fairly tight. I use 0077, but when you need to cpan some common perl modules, switch to the more common 0022.
jobs, fg and bg. Old-school unix commands to play with jobs that.. might take a while. When you forgot to & your command, use CTRL+Z then bg your job.
There may be more, but I can't think of any others right now...
Slashdot Mirror
I'm not sure what you're getting at with this question, but I don't think developing this system, or this kind of thinking in general, is bad.
We slam agencies (Government, FAA) when they are not proactive enough in protecting the general public, as we accuse them of implementing change only in response to bloodshed when a major tragedy occurs. Are we also going to complain when someone is forward-looking into protecting the public against something that has never happened (yet), but could?
The thought of a plane full of people being shot down by an RPG is not that wild and stupid. Do we need to wait for it to happen before looking into solutions? Of course it will cost billions. But can you put a price on your [daughter|son|mother|father|loved one] who could just be sitting on that plane?
I resent that. Just because I look drunk doesn't mean I don't look good.
D.
Here are the top-30 browser stats for www.eclipse.org for September 2006, for those who like to do data mining and stuff like that:
.NET .NET CLR 1 .NET CLR 1 .NET .NET CLR 1 .NET CLR 2
1 21997132 22.37% Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; SV1;
2 9182291 9.34% Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; SV1)
3 4720279 4.80% Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.0.6) G
4 4329042 4.40% Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.0.7) G
5 2555581 2.60% Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; SV1; Maxth
6 2325491 2.36% Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.0;
7 2026769 2.06% Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.0)
8 2026738 2.06% Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1)
9 1873992 1.91% Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; SV1; InfoP
10 1592208 1.62% Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1;
11 1509636 1.54% Java/1.5.0_06
12 1195816 1.22% Mozilla/5.0 (Windows; U; Windows NT 5.1; de; rv:1.8.0.6) Geck
13 1043581 1.06% Mozilla/5.0 (Windows; U; Windows NT 5.1; de; rv:1.8.0.7) Geck
14 1040556 1.06% Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.2; SV1;
15 962330 0.98% Mozilla/5.0 (compatible; Yahoo! Slurp; http://help.yahoo.com/
16 961396 0.98% Java/1.5.0_08
17 810918 0.82% Mozilla/5.0 (compatible; Googlebot/2.1; +http://www.google.co
18 770820 0.78% Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1;
19 747920 0.76% Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.8.0.5) Gecko/200
20 608758 0.62% Mozilla/5.0 (Windows; U; Windows NT 5.1; fr; rv:1.8.0.6) Geck
21 608530 0.62% Java/1.5.0_07
22 599659 0.61% Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.8.0.6) Gecko/200
23 593452 0.60% Mozilla/5.0 (Windows; U; Windows NT 5.1; fr; rv:1.8.0.7) Geck
24 482839 0.49% Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.8.0.7) Gecko/200
25 452618 0.46% Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.7.12) Ge
26 430046 0.44% MnoGoSearch/3.2.34
27 387054 0.39% Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.8.0.6) G
28 362495 0.37% Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.0;
29 343279 0.35% Mozilla/5.0 (Macintosh; U; PPC Mac OS X Mach-O; en-US; rv:1.8
30 341865 0.35% Opera/9.01 (Windows NT 5.1; U; en)
I don't think it's a sense of distrust... But then again, when I got my license to drive at 17, my
parents only very reluctantly lent me their car twice. Despite being on my best behaviour both times, they made it clear that if I wanted to drive, I would have to buy my own car.
So I got a part time job, borrowed money from Dad and bought a decent used car, and paid back every penny of my loan. And hey, the car was mine, so I was free to do as I please. Abuse it? Break it? Too bad, I had to fix it.
Despite that, I still managed to earn a college diploma and today, at 33, I have a wife, kids, 3 cars, a nice house that's paid for and a great job.
I think letting your kids drive around in a vehicle that doesn't belong to them reduces their incentive to work and pay for what they want. Has nothing to do with responsibility. You want it, earn it!
I thought the game was called "Gears of War Ships".. I somehow managed to imagine a game based on a transmission gearbox.
Bah, they'll think of something. They always do. I'll be 65 then, so time won't matter when I'm retired.
Today is 1154536012. None of this mm/dd/yy bullcrap.
Actually, McDonalds doesn't claim to have served "people" - they simply claim "xx billions served". In this case, I think it's more logical to assume they're refering to orders.
As long as Mozilla do like McDonalds - claim "millions of downloads", not "millions of people downloaded" they'll be okay.
Very good question. Sometimes people have a way of taking the fun out of something you do for people.
I use SATA on our smaller, non-mission-critical servers. For our data backend, I wouldn't touch it with a 10-foot pole.
Here are some scenarios where I wouldn't hesitate to use SATA:
- You have redundant servers. Using LVS and/or Heartbeat and your favorite tools, you can get full server redundancy using less expensive hardware. The overall solution can be quite elegant, with hot failover. Why just cover the drives?
- Front-end cluster nodes. You have a powerful, expensive backend server (with a cheaper failover) and you use inexpensive front-end servers for serving client requests. Sounds like overkill for what you want, but with the right server load balancing technology, it can give you a scalable, fault-tolerant and damn fast solution.
- You can live with downtime. Install a server with a couple of SATA disks in a RAID configuration and hope for the best.
I AdBlocked it just before I was tempted to click on it. Bastards won't get me that easily.
Come on. It just *has* to be a joke.
the 12,000 RPM motor is a 60 CC weedwacker motor.
You be wackin' some big weeds with that 60cc weedwacker motor.
I agree, and the tagline is "Slashdot - news for nerds, stuff that matters."
- so it's not a news site, it's a blog
- grammar and spelling seem to matter to the community, but not to the editors
I suggest changing it to "Slashdot - CmdrTaco's blog."
It is your right to tell me to "screw off" if I suggest an improvement - after all, I am free to take my dollars somewhere else if I don't like your site (hereinafter "Product"). Having said that, I like your Product, but I'd like to offer my opinion:
my most-important-link rule
I agree with you.
Next is proper anchor texting.
I agree with you.
I should not be allowed to post my views. I consider this opinion to be simply ludicrous.
I agree. Again, if people don't like it, your readership will go down. Right now your Product is wildly popular, so this is not an issue.
spelling and grammar. Let me be clear. As you are probably well aware, I don't think these are as important as the things I mentioned above.
I disagree with you. Your Product consists of one-paragraph stories, and you only post 50 in a day - they deserve to be properly written and spell-checked. The way I see it, so many people are asking for better spelling, and it would result in a better Product. Are we all wrong? You are under no obligation to improve your Product, but in today's competitive world, if a better Product comes along, you may lose customers. That's up to you.
doesn't really matter much. People found other things to complain about.
Yes, some folks are annoying. I tend to take it with a grain of salt: if one person complains about everything about your Product, then your Product is not for them. If everyone complains about the same thing, then ignoring those complaints is being short-sighted (or close-minded).
We get millions of e-mails a day, and everyone complains
We don't care - quit your whining. Every job on this planet is hard. You are offering a Product, and if delivering said Product is too much work, you have many options:
a) quit
b) make processes more efficient
c) hire more staff
d) delegate: levrage the power of your community. Have the community edit articles. You can make it work. What? No money to modify Slashcode? Have call for contributions. A code camp to improve Slashcode. A fundraising campaign. I'd gladly contribute $10 towards making Slashdot a more community-driven site. Right now.
D.
it's always not enough for SOMEONE
I've been coming here for eight years, and Taco, you *still* seem to take criticism too personally. I was accused of that not long ago. When something's your baby and it's popular, you tend to focus on the negative.
You can't please everybody so quit trying. Enjoy what you're doing for the millions who like it, who use it and who don't say a word. But...
Don't hesitate to listen to those who say something's broken. If one person sais it's broken, laugh - it's funny. When one hundred say it's broken, suck it up - it most likely is. BUT that doesn't mean it's gawd-awful and it's certainly NOT a mandate to fix it. If you don't have the resources to make it better, so be it. Don't beat yourself up. Add it to the FAQ.
Be more transparent. If we knew what your workday looked like, maybe we'd understand. As it is, we don't.
Be fair with yourself, and with your readers. Like above, many bitch and complain, but we're not always wrong. Likewise, we don't know the whole story behind Slashdot, so keep that in mind. When someone sais you should post reject notices to stories, they have no clue what that involves. Chuckle and add it to the FAQ. Life sucks for them.
D.
A sample size of 1 is adequate when you only have one server - and you run the benchmark on the actual server you're going to be using.
Having a benchmark use a sample of 1000 different units and tell you that ReiserFS 4 is the best is useless if it performs worst on your specific hardware platform.
Here's a modding tip: go ahead and buy the Raptor drive, and mod it by removing the clear cover to install a solid, opaque one!
An interesting analysis in every aspect, and it's fine and dandy for the person who uses 400 GB drives and a ATA controller on a 500MHz computer but I'd like to see how the filesystems compare on a bigass RAID system run by a Power5 server, or a few Itaniums that usually have with a few hundred connected users. Something a bit more "entreprise" - where the choice of a filesystem is a bit more critical than a small server or a home PC.
*applause*
Back in the day, I think the only reason Novell tacked on a GUI to NetWare was because of the pressure they were feeling from Windows NT and the new wave of "Admins" who were addicted to the mouse. NetWare 3.12/4.x was *the* file and print solution, and NDS is just so many miles ahead of Active Directory.
We run a cluster of SLES 9 servers, consisting of 4 Itanium2's, 4 x86_64's, 3 IBM Power5's and one x86, and they run flawlessly. No X, no KDE crap, just a barebones minimal install + the 25-or-so server packages we need. Keeping the OS up-to-date has been flawless as well, including Kernel upgrades.
It's nice to have a unified OS for all the hardware platforms we run.
To me, Novell rocks.
Here's what I think he'll answer to your first question:
"
MySQL works fine on Slashdot. It has all the features and performance we need, it has been running flawlessly for years and we're already familiar with it, so why should we change to anything else? What makes you think there's a bullet that needs biting? Granted, PG looks neat and all, but why exchange a dollar for four quarters?
"
Here's what I think he'll answer to your second question:
"
MySQL 5 doesn't offer us any features we absolutely need (otherwise we'd be using PG, right?) We will upgrade eventually, but we have bigger fish to fry right now, and upgrading our database is not very imperative.
"
Something else he might say:
"
Running slashdot is not as simple as running a basement website that gets 3 hits per hour. Thought needs to be put into these decisions. We can't just run off and install something the day it's released.
"
Applause.
ethtool and mii-tool. The Cisco 2970 switch we use has the knack of initializing the interface in half-duplex mode if the port is set to "Auto". Easily fixed by setting the port to Full, but useful nonetheless.
/etc/hosts.allow and /etc/hosts.deny, or where pop3/imap are looking for PEM certificates to configure pop3s/imaps, etc...
/proc filesystem. Favorites include /proc/net/dev, and /proc/uptime, /proc/cpuinfo, /proc/loadavg. Good for aggregating individual server load data in a cluster.
... permissive. Cut down on access to your logfiles and to config files in /etc.
.. might take a while. When you forgot to & your command, use CTRL+Z then bg your job.
strings. Good to check if executables are using
vmstat. Think your system is paging, or a card is generating too many interrupts?
awk and sed. Mentioned elsewhere, but priceless.
chmod. I think the Linux filesystem permissions are too
*quota*. A must for restricting disk space use.
umask. When you need root, set your default umask fairly tight. I use 0077, but when you need to cpan some common perl modules, switch to the more common 0022.
jobs, fg and bg. Old-school unix commands to play with jobs that
There may be more, but I can't think of any others right now...
You need to sort first so that uniq sees the repetition all in one block, then sort after for the numerical order.
.bash_history | awk '{print $1}' | sort | uniq -c | sort -nr | head -20
cat
cat .bash_history | awk '{print $1}' | sort | uniq -c | sort -nr | head -20