Netcraft #'s are useful for decision making...
on
Apache down, IIS up
·
· Score: 1
It's interesting to speculate about the Netcraft #'s, but really, people spend too much time worrying about it. Who cares what a million tiny sites use as a webhost? One busy site can easily serve more people than 10's of thousands of smaller sites. The Netcraft survey is just a fun, easy to automate, bit of research. The trends are somewhat interesting, but kind of useless if a single registrar moving parked domains can move the numbers.
Well, whatever. I haven't seen much Windows 9x for the last couple years now. Of the remaining Windows 9x machines, how many can actually run Firefox usably anyway? Firefox is pretty happily sucking up over 100MB of memory on my laptop right now, most Windows 98 era machines would have maybe 64MB?
Further, what about old Linux distros? What are the odds of Firefox 3 running on RedHat 5?
The PR blurb is a little light on the details. Does anyone know if there will be speed benefits (or, IMHO, less likely power benefits) for existing laptops? I.e. should I look forward to giving my laptop a bit of a boost with one of these drives? I know that Vista is supposed to have a lot of code to really benefit from hybrid drives... but I imagine that at least some benefits might be available to XP or Linux.
Does anyone smarter than me know more about these drives?
I suspect that Sony will keep the hardware PS2 for "emulation" for the life of the PS3. According to my 2 minutes of research, the Emotion Engine "only" has 10.5 million transistors (at least that's what the wikipedia emotion engine page states. This is less than the # of transistors in a single SPE of the PS3's Cell. The Cell processor has 7 SPE's, plus the PowerPC core, and an 8th disabled or "spare" SPE. Again, the wiki link.
Assuming that both transitor counts are comparable and reasonably accurate, I don't see this being a problem in the long run. The "emulation" waste ratio looks like less than 10%, which I'm guessing is comparable to the PS1 in the PS2. Since Sony has a history of providing backwards compatibility, a few more bucks on a separate chip today and a bit of wasted wafer space in the future (after integration into a single chip), I think that a hardware emulator is the way it will be.
Why do a few domains imply malware???
on
Predicting Malware
·
· Score: 2, Insightful
IMHO, the far more likely purpose of registering domains related to the next hurricane names are simply for SPAM. When the hurricanes hit, tens of thousands of people will mistype or find some other way of ending up on what I expect will be plain old ad serving pages. Considering the miniscule costs of setting this up, I'd suspect that it would make a few bucks. Especially compared to some of the other ad based domains I've stumbled on in the past...
Off the top of my head, I can recall a few events that definitely have slowed the Internet overall -- maybe not "meltdown" but definitely noticeable and bothersome. Let's see, Victoria's Secret streaming their big show back in 2000 or so, the SQL Server Blaster worm, and 9/11 (though, that was more specific to individual websites than the 'net as a whole, there was the big peering conflict of last year.... overall, the Internet isn't quite as robust as advertised. I'm sure that the military has their own "internet like" system for communications in emergencies, because I'm sure they wouldn't rely on the Internet as it stands today. Heck, they've probably got a half dozen plans on ways to break the Internet so that the enemy of the day can't use it at critical times -- just like GPS.
However, with regards to the World Cup, here in North America where most people think football involves using hands to throw and catch a non-round ball, I don't think we'll notice the traffic...:)
When stuff breaks unexpectantly (like Apollo 13) and you have to jerry-rig something in space, it's cool to DIY. But to loose a tool because it's "just like loosing your car keys", then it's not so smart. Those tools aren't cheap, and the gear up there that is at risk from a missing tool is far far far from cheap.
So... maybe this would be a really good application for RFIDs? Stick a tag on all the tools and at least you'll have an easy way to find anything that get's lost.
there must be something in the air. First the Yahoo-Ebay thing (which turned out to be real), then the AMD-ATI thing, and now an Apple-RIM thing? What's next? A MS-Nokia deal?
This Summer of Code program of Google's is brilliant. I'm amazed at how every major OSS project has gotten funding for at least a couple projects. Maybe it's time for other firms to start doing something similar, the Cottage Fever Coding program(?)... it would certainly be good for PR.
The "problem" with encryption is not that it takes a few extra clicks, a little extra training and a more effort to work with third parties. Though, those are issues. To me, the real problem is that we're lazy and there is no benefit... at least until something bad happens.
I've recently been part of a outsourced payroll provider transition. As the IT guy for the customer end, I've been telling our folks to put a password on sensitive Excel* files (full of SIN's and other stuff like that) that get emailed to the payroll company. I initiated that, not the "professional" payroll firm... why don't they insist on it? Who knows? I would guess that it just causes too many questions and minor problems. As long as people don't screw up, there is no "need" for encryption. Simple as that.
* Yes, I know, default Excel encryption isn't that secure, but it's good enough to guard against mistyped email addresses, and it's hard to beat the simplicity and ubiquity. Good luck sending a TrueCrypt file to someone outside the company!
AGP was not very useful for bidirectional data flow, but PCIe is. GPU's are pretty sophisticated these days, so they've got the logic to handle moving stuff in and out of it's memory and over the bus to the CPU and the rest of the system.
I've visited paypalsucks.com in the past... sure, you'll find a few unhappy PayPal users. What you won't see are the millions (probably 10's of millions) of people who are happy with PayPal -- including me.
There are complaints about various Google products too. It's just that most people ignore them since Google will "do no evil". IMHO, keep your eyes on click fraud being a bigger issue in the near future. Besides, it's in the best interest of traditional advertising providers to make it a big deal since it'll make traditional ads appear safer.
It matters to many of their employees. Valuable stock options are a great way to motivate people. Worthless stock options are pretty demoralizing.
Also, if the price of their stock did drop too much, they would run the risk of an unfriendly takeover, unless of course the insiders still have most of the shares -- I'm not sure if they do or not. Either way, the stock would really have to tank for that to be likely.
Anyway, it's all fine for Google to pretend to be above the Wall Street games when the times are good. Somewhere down the road though, revenues will level off, competition will catch up, and they're going to wish they didn't piss off all the financial folks.
Dell is probably going to get a lot of incremental profits from this deal! I would assume that the deal is comparable to the one that Mozilla receives for the Firefox default search engine box, i.e. some fraction of the paid advertising revenue from searches on Google.
Let's see, the Mozilla Foundation received something in the 7 figure territory last year from Google. Since Dell sells way more desktops than Firefox has users, plus this deal (probably) includes the Google toolbar which is far more valuable to Google than just being the default search engine, I wouldn't be surprised if this was a worth a couple hundred million dollars a year for both companies.
Not crazy huge money for companies the size of Google and Dell, but pretty good! Granted, we don't know what Yahoo or MSN would offer Dell as an alternative... so potentially Google is benefiting more from this deal, maybe.
Especially at the low end, you have to do a bit of searching on sites like SitePoint forums or WebHostingTalk. Keep in mind, almost every large web host will get a bad review now and then; however, as long as there doesn't seem to be a large number of complaints, you're off to a good start. Unfortunately, with hosting, it is possible to offer pretty decent basic hosting packages at very low prices. If the host is good everything is automated, hardware is cheap (for everyone), and bandwidth at a datacenter is cheap. So it's not safe to assume that you get what you pay for.
The best thing to do IMHO, is to get a personal reference from either a comparable website to yours or a friend who actually uses the product. For instance, when I was looking at VPS's about two years ago, it was very reassuring to find out that Damn Small Linux was hosted with the provider that I had short-listed (and eventually chose).
Uh, whoop-de-doo. I can't buy games for my old Sega Master System either... what's the big surprise again? XBox had a reasonably long life for a console. Let it be.
As long as you are aware of the limitations, you'll be fine. For small scale use, the only one item that I would consider is whether or not you want RAID for hard disk storage -- you'll need to use external drives if you do. Otherwise the laptops are pretty much regular computers, albeit slightly slower, as far as these things go. Oh, and you won't have fancy remote management tools that medium and higher end servers have.
Are there any OSS distros that are specifically made for scanning email for viruses and flagging spam? Something akin to say Smoothwall or IpCop for firewalls.
At work we have a 30 seat license to SAV w/server based email scanning. I'd happily switch to something cheaper than SAV; however, once I price in the server based email scanning, there hasn't been much savings in the past. The email scanning is pretty much half the cost, but it is something that could be done well by an OSS disto. I am NOT willing to go dinking around with more than a handful of config files and packages to get this working, however. At some point it's just not worth the time.
There's a lot of chatter in this thread about this being little more than a PR stunt with limited sales value to AMD. At face value, this is true; however, consider the bigger picture. This news validates AMD's product -- Dell saying that a product is good is much better for AMD than Dell saying that "there's no demand" as was the case not that long ago. Remember, Dell is using this for their most expensive servers -- that's nothing to sneeze at. Heck, the profit for AMD for 4x800 series Opterons is the probably close to the profit that Intel gets from 100 Celerons. AMD probably has a gross margin of close to $1000 on those 800 series Opterons, versus maybe $20-50 for the Celerons.
Dell's not announcing anything lower than the four cpu servers, but given the situation, Dell has no choice but to take baby steps. We're talking about huge volumes of chips in Dell's mainstream lines, so a little caution is reasonable! Nevertheless, now that Dell uses AMD chips, the next guessing game becomes "How long until their customers force Dell to get AMD 2-way servers (or maybe 64bit laptops)?". I'm thinking less than a year. They've already taken the hardest step of ditching the exclusivity with Intel. There's no real reason to hold back now.
Great! I've wanted to use SkypeOut for quite a while. However, I just couldn't get their website to take my credit card. (Or maybe my CC company was auto rejecting the request...) Anyway, it wasn't worth the effort to figure out what was wrong just to save the extra $.01/min (vs. a calling card) on North American long distance (which I was most interested in)...
Actually, there was no modification required at all! Just pop in the properly prepared CDR. There was no risk to my warranty. No messy soldering to screw up. And no risk of bricking my Dreamcast... or even if I did, it'd be "covered" by the warranty. This hack for the X360 is not in the same league as the Dreamcast hack.
It's my rough understanding that the bulk of SOX is to add a whole lot of layers to make sure the numbers balance and that you know who officially made what changes to numbers.
Enron was about fraud by management. Fraud will find a way to trump any set of rules to the contrary -- except maybe moral and ethical, but those are pretty hard to quantify.
Will SOX make it harder for another Enron to occur? Yeah, maybe. Will it prevent large scale corporate collapses from fraud? Not a chance. There will always be new and ingenious ways to break the system. Never mind the legal ways for management to line their pockets. Really, you'd think a couple million a year would be enough for anyone...
I'm not so old as to have completely forgotten university life... IMHO, student labs aren't that great an environment for thin clients. (Not that there are that many great situations in general!)
Disclaimer: I've only been responsible for a small Windows TS install - half a dozen clients over the 'net... so take this with a large grain of salt.
Here's a simple test. Turn on a basic CPU usage chart (e.g. the ubiquious Task Manager in Windows). Browse the web and do basic run of the mill stuff that a university kid would do. Remember, young folks are perfectly capable of handling a half-dozen browser windows at once. Every time that CPU chart spikes, you run the risk of a slow down for everyone on the same server. True, it's not totally comparable, but then again, you'll have 20+ users on the same machine. Every student lab needs full access to the web, including Flash, Java, and even DHTML (ok, "AJAX"). It doesn't take long before you run out of all the processing capacity that you have available. Oh, and media is becoming more and more a core part of using the internet. That's torture on thin clients.
For managability, things are way better than they used to be in the world of Windows. Disk imaging software is better and cheaper. Limited access accounts are easier to setup and most newer applications are (finally) compatible with that access level. And hardware is dirt cheap -- cheaper than the comparatively low-volume thin clients -- if you get selective with CPU's, the power difference doesn't have to be that large.
Plus, in the end, you'll have vastly more CPU and hard drive space "around" the network. Sooner or later someone will find something smart to do with it all... like distributed backups or something.
Though, you'll have to avoid the fancy-shmancy latte's and mocha's. A regular cup of Joe and a teaspoon or two of sugar give me all the caffeine I need -- at least until the next cup. =)
Slashdot Mirror
It's interesting to speculate about the Netcraft #'s, but really, people spend too much time worrying about it. Who cares what a million tiny sites use as a webhost? One busy site can easily serve more people than 10's of thousands of smaller sites. The Netcraft survey is just a fun, easy to automate, bit of research. The trends are somewhat interesting, but kind of useless if a single registrar moving parked domains can move the numbers.
Well, whatever. I haven't seen much Windows 9x for the last couple years now. Of the remaining Windows 9x machines, how many can actually run Firefox usably anyway? Firefox is pretty happily sucking up over 100MB of memory on my laptop right now, most Windows 98 era machines would have maybe 64MB?
Further, what about old Linux distros? What are the odds of Firefox 3 running on RedHat 5?
The PR blurb is a little light on the details. Does anyone know if there will be speed benefits (or, IMHO, less likely power benefits) for existing laptops? I.e. should I look forward to giving my laptop a bit of a boost with one of these drives? I know that Vista is supposed to have a lot of code to really benefit from hybrid drives... but I imagine that at least some benefits might be available to XP or Linux.
Does anyone smarter than me know more about these drives?
I suspect that Sony will keep the hardware PS2 for "emulation" for the life of the PS3. According to my 2 minutes of research, the Emotion Engine "only" has 10.5 million transistors (at least that's what the wikipedia emotion engine page states. This is less than the # of transistors in a single SPE of the PS3's Cell. The Cell processor has 7 SPE's, plus the PowerPC core, and an 8th disabled or "spare" SPE. Again, the wiki link.
Assuming that both transitor counts are comparable and reasonably accurate, I don't see this being a problem in the long run. The "emulation" waste ratio looks like less than 10%, which I'm guessing is comparable to the PS1 in the PS2. Since Sony has a history of providing backwards compatibility, a few more bucks on a separate chip today and a bit of wasted wafer space in the future (after integration into a single chip), I think that a hardware emulator is the way it will be.
IMHO, the far more likely purpose of registering domains related to the next hurricane names are simply for SPAM. When the hurricanes hit, tens of thousands of people will mistype or find some other way of ending up on what I expect will be plain old ad serving pages. Considering the miniscule costs of setting this up, I'd suspect that it would make a few bucks. Especially compared to some of the other ad based domains I've stumbled on in the past...
Off the top of my head, I can recall a few events that definitely have slowed the Internet overall -- maybe not "meltdown" but definitely noticeable and bothersome. Let's see, Victoria's Secret streaming their big show back in 2000 or so, the SQL Server Blaster worm, and 9/11 (though, that was more specific to individual websites than the 'net as a whole, there was the big peering conflict of last year.... overall, the Internet isn't quite as robust as advertised. I'm sure that the military has their own "internet like" system for communications in emergencies, because I'm sure they wouldn't rely on the Internet as it stands today. Heck, they've probably got a half dozen plans on ways to break the Internet so that the enemy of the day can't use it at critical times -- just like GPS.
:)
However, with regards to the World Cup, here in North America where most people think football involves using hands to throw and catch a non-round ball, I don't think we'll notice the traffic...
When stuff breaks unexpectantly (like Apollo 13) and you have to jerry-rig something in space, it's cool to DIY. But to loose a tool because it's "just like loosing your car keys", then it's not so smart. Those tools aren't cheap, and the gear up there that is at risk from a missing tool is far far far from cheap.
So... maybe this would be a really good application for RFIDs? Stick a tag on all the tools and at least you'll have an easy way to find anything that get's lost.
there must be something in the air. First the Yahoo-Ebay thing (which turned out to be real), then the AMD-ATI thing, and now an Apple-RIM thing? What's next? A MS-Nokia deal?
Let's speculate like it's 1999!
This Summer of Code program of Google's is brilliant. I'm amazed at how every major OSS project has gotten funding for at least a couple projects. Maybe it's time for other firms to start doing something similar, the Cottage Fever Coding program(?)... it would certainly be good for PR.
The "problem" with encryption is not that it takes a few extra clicks, a little extra training and a more effort to work with third parties. Though, those are issues. To me, the real problem is that we're lazy and there is no benefit... at least until something bad happens.
I've recently been part of a outsourced payroll provider transition. As the IT guy for the customer end, I've been telling our folks to put a password on sensitive Excel* files (full of SIN's and other stuff like that) that get emailed to the payroll company. I initiated that, not the "professional" payroll firm... why don't they insist on it? Who knows? I would guess that it just causes too many questions and minor problems. As long as people don't screw up, there is no "need" for encryption. Simple as that.
* Yes, I know, default Excel encryption isn't that secure, but it's good enough to guard against mistyped email addresses, and it's hard to beat the simplicity and ubiquity. Good luck sending a TrueCrypt file to someone outside the company!
AGP was not very useful for bidirectional data flow, but PCIe is. GPU's are pretty sophisticated these days, so they've got the logic to handle moving stuff in and out of it's memory and over the bus to the CPU and the rest of the system.
I've visited paypalsucks.com in the past... sure, you'll find a few unhappy PayPal users. What you won't see are the millions (probably 10's of millions) of people who are happy with PayPal -- including me.
There are complaints about various Google products too. It's just that most people ignore them since Google will "do no evil". IMHO, keep your eyes on click fraud being a bigger issue in the near future. Besides, it's in the best interest of traditional advertising providers to make it a big deal since it'll make traditional ads appear safer.
It matters to many of their employees. Valuable stock options are a great way to motivate people. Worthless stock options are pretty demoralizing.
Also, if the price of their stock did drop too much, they would run the risk of an unfriendly takeover, unless of course the insiders still have most of the shares -- I'm not sure if they do or not. Either way, the stock would really have to tank for that to be likely.
Anyway, it's all fine for Google to pretend to be above the Wall Street games when the times are good. Somewhere down the road though, revenues will level off, competition will catch up, and they're going to wish they didn't piss off all the financial folks.
Dell is probably going to get a lot of incremental profits from this deal! I would assume that the deal is comparable to the one that Mozilla receives for the Firefox default search engine box, i.e. some fraction of the paid advertising revenue from searches on Google.
Let's see, the Mozilla Foundation received something in the 7 figure territory last year from Google. Since Dell sells way more desktops than Firefox has users, plus this deal (probably) includes the Google toolbar which is far more valuable to Google than just being the default search engine, I wouldn't be surprised if this was a worth a couple hundred million dollars a year for both companies.
Not crazy huge money for companies the size of Google and Dell, but pretty good! Granted, we don't know what Yahoo or MSN would offer Dell as an alternative... so potentially Google is benefiting more from this deal, maybe.
Especially at the low end, you have to do a bit of searching on sites like SitePoint forums or WebHostingTalk. Keep in mind, almost every large web host will get a bad review now and then; however, as long as there doesn't seem to be a large number of complaints, you're off to a good start. Unfortunately, with hosting, it is possible to offer pretty decent basic hosting packages at very low prices. If the host is good everything is automated, hardware is cheap (for everyone), and bandwidth at a datacenter is cheap. So it's not safe to assume that you get what you pay for.
The best thing to do IMHO, is to get a personal reference from either a comparable website to yours or a friend who actually uses the product. For instance, when I was looking at VPS's about two years ago, it was very reassuring to find out that Damn Small Linux was hosted with the provider that I had short-listed (and eventually chose).
Uh, whoop-de-doo. I can't buy games for my old Sega Master System either... what's the big surprise again? XBox had a reasonably long life for a console. Let it be.
As long as you are aware of the limitations, you'll be fine. For small scale use, the only one item that I would consider is whether or not you want RAID for hard disk storage -- you'll need to use external drives if you do. Otherwise the laptops are pretty much regular computers, albeit slightly slower, as far as these things go. Oh, and you won't have fancy remote management tools that medium and higher end servers have.
Are there any OSS distros that are specifically made for scanning email for viruses and flagging spam? Something akin to say Smoothwall or IpCop for firewalls.
At work we have a 30 seat license to SAV w/server based email scanning. I'd happily switch to something cheaper than SAV; however, once I price in the server based email scanning, there hasn't been much savings in the past. The email scanning is pretty much half the cost, but it is something that could be done well by an OSS disto. I am NOT willing to go dinking around with more than a handful of config files and packages to get this working, however. At some point it's just not worth the time.
There's a lot of chatter in this thread about this being little more than a PR stunt with limited sales value to AMD. At face value, this is true; however, consider the bigger picture. This news validates AMD's product -- Dell saying that a product is good is much better for AMD than Dell saying that "there's no demand" as was the case not that long ago. Remember, Dell is using this for their most expensive servers -- that's nothing to sneeze at. Heck, the profit for AMD for 4x800 series Opterons is the probably close to the profit that Intel gets from 100 Celerons. AMD probably has a gross margin of close to $1000 on those 800 series Opterons, versus maybe $20-50 for the Celerons.
Dell's not announcing anything lower than the four cpu servers, but given the situation, Dell has no choice but to take baby steps. We're talking about huge volumes of chips in Dell's mainstream lines, so a little caution is reasonable! Nevertheless, now that Dell uses AMD chips, the next guessing game becomes "How long until their customers force Dell to get AMD 2-way servers (or maybe 64bit laptops)?". I'm thinking less than a year. They've already taken the hardest step of ditching the exclusivity with Intel. There's no real reason to hold back now.
Great! I've wanted to use SkypeOut for quite a while. However, I just couldn't get their website to take my credit card. (Or maybe my CC company was auto rejecting the request...) Anyway, it wasn't worth the effort to figure out what was wrong just to save the extra $.01/min (vs. a calling card) on North American long distance (which I was most interested in)...
Actually, there was no modification required at all! Just pop in the properly prepared CDR. There was no risk to my warranty. No messy soldering to screw up. And no risk of bricking my Dreamcast... or even if I did, it'd be "covered" by the warranty. This hack for the X360 is not in the same league as the Dreamcast hack.
It's my rough understanding that the bulk of SOX is to add a whole lot of layers to make sure the numbers balance and that you know who officially made what changes to numbers.
Enron was about fraud by management. Fraud will find a way to trump any set of rules to the contrary -- except maybe moral and ethical, but those are pretty hard to quantify.
Will SOX make it harder for another Enron to occur? Yeah, maybe. Will it prevent large scale corporate collapses from fraud? Not a chance. There will always be new and ingenious ways to break the system. Never mind the legal ways for management to line their pockets. Really, you'd think a couple million a year would be enough for anyone...
Do'h. Forgot the html formatting... blah.
I'm not so old as to have completely forgotten university life... IMHO, student labs aren't that great an environment for thin clients. (Not that there are that many great situations in general!) Disclaimer: I've only been responsible for a small Windows TS install - half a dozen clients over the 'net... so take this with a large grain of salt. Here's a simple test. Turn on a basic CPU usage chart (e.g. the ubiquious Task Manager in Windows). Browse the web and do basic run of the mill stuff that a university kid would do. Remember, young folks are perfectly capable of handling a half-dozen browser windows at once. Every time that CPU chart spikes, you run the risk of a slow down for everyone on the same server. True, it's not totally comparable, but then again, you'll have 20+ users on the same machine. Every student lab needs full access to the web, including Flash, Java, and even DHTML (ok, "AJAX"). It doesn't take long before you run out of all the processing capacity that you have available. Oh, and media is becoming more and more a core part of using the internet. That's torture on thin clients. For managability, things are way better than they used to be in the world of Windows. Disk imaging software is better and cheaper. Limited access accounts are easier to setup and most newer applications are (finally) compatible with that access level. And hardware is dirt cheap -- cheaper than the comparatively low-volume thin clients -- if you get selective with CPU's, the power difference doesn't have to be that large. Plus, in the end, you'll have vastly more CPU and hard drive space "around" the network. Sooner or later someone will find something smart to do with it all... like distributed backups or something.
Though, you'll have to avoid the fancy-shmancy latte's and mocha's. A regular cup of Joe and a teaspoon or two of sugar give me all the caffeine I need -- at least until the next cup. =)