Purists would even disagree with the article author's implied definition of Computer Science, i.e. when she mentions "program design". For some folks "Computer Science" is wholly separate from "programming".
Break up companies such that there are "pipe providers" and "content providers". If my net service provider is not also my cable TV company then it loses much of its motivation to oppose net neutrality. Of course, this may not go far enough. If I'm an ISP and my backbone connection doesn't provide sufficient bandwidth to Comcast's digital cable stream, Comcast will be unhappy (and so will my users). So I'll need to add that bandwidth. The question is how much of that effort should be paid for by me, the "pipe guy" and Comcast (or any other content provider). I'm tempted to say that the market would dictate "pipe provider" behavior without any direct intervention from content providers. For instance, if Comcast offers digital cable and says "our service works on X, Y and Z providers, but not U, V and W", then U/V/W are going to face pressure to properly support Comcast's content stream.
I recently bought several home appliances (washer, dryer, gas oven) at a local appliance store (Conn's). After I'd picked out what I wanted, the store guy went to a computer and looked up web prices on the same items at Best Buy, Lowe's, Sears and Home Depot. Then he knocked about $200 to bring it in line with what one of the other places was offering.
The team planted four types of genes into human skin cells to create the iPS cells, according to Kyodo News.
...which is then followed by this:
Scientists say the use of human embryonic stem cells as a treatment for cancer and other diseases holds great promise...
Here's what confuses me: the first bit seems to suggest the stem cells used to "make the monkey jump" were adult, not embryonic. So why include the last little bit about embryonic stem cell research? Am I incorrect about the first quote, and in fact the cells they used were embryonic in origin?
Point taken. But I also think this approach would change the way people conduct their lives and the way in which local governments operate so as to make both much more expensive and less efficient. If the goal weren't so much "terror" as it was "requiring the addition of security 'overhead' to anything and everything", then this approach would do the trick.
I've always thought terrorists could get a lot more bang for their buck by using much less extravagant means. You don't need to crash a jet into a building; just replicate the D.C. sniper from a few years ago across 50 different U.S. cities. Target local government officials, police officers, women, children, etc. All you need is to get some guys with marksmanship training into the country then get some high-powered rifles into their hands. Instruct them to take their own lives if capture is imminent. I have a feeling this tactic would go a lot further towards "instilling fear" in the U.S. populace on a day-to-day basis than the plane crashing thing.
For such a small organization you might want to keep things as simple as possible given you're not going to have many staff to support 20 employees. Probably "one guy". To that end:
* Branded Gmail for email and calendar. You can use the branded google accounts for IM as well. The spam filtering and uptime are very good. Also you don't have to manage any of it. Your employees can automatically check their work email from anywhere w/o having to get on a VPN or use a particular email client.
* Macs. Makes you functionally immune to malware. Repairs/replacement are pretty speedy, esp. if you have a Mac store in your area. If you absolutely positively must have MS Office then you can get it as a native app. If you must run Windows then there's a free virtualization option (Virtual Box). It's not as good as Parallels, but it gets the job done.
* If you expect your employees to occasionally work from home (or on the road) or if you want to at least give them that option, then get everyone laptops. Providing external monitors, keyboard, mouse is fairly cheap. MacBooks (not MacBook Pro) are in the same ballpark cost-wise as similarly spec'd name-brand PC laptops.
* Hosting your own web server sounds like an unnecessary pain in the butt. If you absolutely must, then Linux/Apache is probably the way to go. I'd recommend the latest LTS version of Ubuntu (10.04). Going with an esoteric distribution just makes finding documentation and fixing problems that much more time-consuming.
Use it until they make you pay for it or the features become unbearably annoying. Who cares if its from Oracle? Its not like you're spending money on it. Personally I've always found OO to be a giant pile of crap and use MS Word instead. 2003 version; I don't especially like the new 2007 look and feel.
Define "drying up". If you look at the percentage of non-Microsoft web-based development positions based around Java and PHP...its a pretty high percentage. Personally I wouldn't waste my time learning a niche language/framework like RoR. Learning to do Android or iPhone development would probably be useful.
For maximum simplicity just move everyone to OS X. The guys who need Office can use MS Office for the Mac. The non-technical users won't be freaked out by OS X in the way they might be by Linux. OS X gives you most of the same malware immunity you get with any other non-Windows OS. The marketing and graphics guys get to keep using Macs just like they always have. Your developers, if there are any, should be fine on the Mac unless they're doing development that specifically targets Windows.
From a cost perspective, of course, this may not be the cheapest thing in the world. But you can't beat it for simplicity. I also like the branded Gmail suggestion. That can work irrespective of which OS's you deploy.
I think you misunderstand me; generally speaking like OS X. Furthermore, I suspect the fact that it lacks some of the irritations of, say, an Unbuntu, is largely because of Apple's dictatorial control and closed environment. OS X seems to hit a nice sweet spot between non-technical usability and not completely shafting technical users. I appreciate being able to drop to a bash prompt and do unixy things when I need to. I like not ever being forced to do that.
This article seems like a big excuse. If you can't beat 'em then don't try. Want a great example of a Unix-like OS beating MS at their own game? OS X. Know what it took? Some quality building blocks, then dictatorial stewardship of a single company with deep pockets and a willingness to meet the needs of a completely non-technical user.
I disagree with the "idiocracy" tag on this article. It is perfectly rational behavior for me to push the "walk" button at a crosswalk even if there's a decently high chance it will have no effect. Why? Because the cost to me is very small vs. a potentially large payoff if it actually does work.
I love that Phoronix is willing to take the time to run tests like this. I just wish they'd learn how to run meaningful tests. For instance, why are they testing a bunch of CPU-bound things? Kernel won't affect that unless we're talking about SMP performance. If you want to test the kernel, test how well it handles SMP, network I/O and disk I/O. And bear in mind that disk I/O will be hugely affected by which filesystem is used and its configurable settings.
Another problem with their article is that it tests individual kernels. Most folks don't use a vanilla kernel. They use one provided by their distro, which may have distro-specific patches that address some of the performance problems (or add new ones). What I would have preferred to see is a comparison of different distro releases over the last 5 years, focusing on the most popular ones (say Ubuntu, Fedora and SuSE).
The meaningful tests (and their results) were:
1. GnuPG: avoid 2.6.30 and later.
2. Loopback TCP: avoid 2.6.30 and later.
3. Apache Compilation: avoid 2.6.29 and earlier.
4. Apache static content: avoid 2.6.12, 2.6.25, 2.6.26, then 2.6.30 and later.
5. PostMark: avoid 2.6.29 and earlier.
6. FS-Mark: avoid 2.6.17 and earlier, 2.6.29, then 2.6.33 to 2.6.36.
7. ioZone: unless you're willing to run 2.6.21 or earlier, avoid 2.6.29 and you're fine.
8. Threaded I/O: avoid 2.6.20 and earlier, 2.6.29, then 2.6.33 to 2.6.36.
Based on these results, #1 and #2 seem to be testing the same thing, and tests #3 and #5 seem to be testing the inverse of whatever that thing is. 2.6.29 seems to be especially crappy, performing worse than the kernels immediately before and immediately after it on tests #6, #7 and #8. In terms of recent kernels, tests #6 and #8 suggest a regression in 2.6.33 that has been resolved in 2.6.37.
If it were me, I'd look at either running 2.6.37 (when its released) or fall back to 2.6.32 if my hardware was supported.
Yes. I've long maintained that a course in prob/stat would be way more useful than, say, trigonometry and calculus for students who aren't going into science/engineering professions. Honestly if you could just ram home two basic concepts: 1. correlation is not causation, and 2. what "statistical significance" means, then that would serve most folks well.
I was going to post this as well, but I'll just agree with what you wrote. Originally Apple probably rolled its own JRE because it didn't think it had a large enough Java user base to force Sun to target OS X. Apparently Apple now thinks it does have a large enough user base to force Oracle's hand. So they stop rolling their own and let Oracle take the heat if it decides to abandon Mac users.
Though I'd say it was always pretty much dead. Linux on the desktop is fine for, say, a developer workstation. Or maybe something like a netbook or smartphone OS where the content is mostly managed and/or runs in a sandbox that's entirely shielded from the linux environment (e.g. Android). Anything else its less trouble to run Windows or OSX.
Exactly. Here are Microsoft's statistics after computer prevalence is taken into account. Quote from that page:
Among locations with more than 200,000 executions of the MSRT in 2Q10, Turkey had the highest infection rate, with 36.6 computers cleaned for every 1,000 MSRT executions (CCM 36.6). Following Turkey were Spain (35.7), Korea (34.4), Taiwan (33.5), and Brazil (25.8). All have been among the locations with the highest infection rates for several periods.
Locations with the lowest infection rates include Belarus (1.3), Bangladesh (1.5), Sri Lanka (1.8), Tunisia (1.8), and Morocco (1.9).
Given the very low infection rate of most of Africa, though, something tells me Microsoft's "CCM" metric may not perfectly reflect real infection rates.
I worked for a startup that used branded google services for email, instant messaging and calendar. It worked pretty well. Not sure how the cost worked out, but it was certainly less headache than maintaining our own mail server. Especially since we didn't have a dedicated IT person.
2. While its sync functionality does run automatically at browser shutdown, it currently doesn't sync stored passwords.
3. Since there is no Application Bar in XP like there is on OSX I have to configure Chrome to have the "bookmark bar" visible on every tab.
4. On XP Chrome exhibited a UI bug where the horizontal space at the bottom of the screen used to display downloaded files became permanently "blank". I wasn't able tor reproduce it, but the fact that it happened at all was annoying.
5. Scrolling on XP was slow and jumpy compared to Firefox and IE. Not sure what's going on there. I tested it side-by-side w/ Firefox so I know it's not my mouse or mouse driver.
6. While Chrome is faster on JS-intensive tasks it "feels" slower on regular page loads. My suspicion is that this is due to Chrome starting to render the page before all elements are received, which causes the page to visibly change as new elements are loaded. Firefox has a short delay where there are no visible changes then displays the page all at once. For whatever reason I find this more pleasing.
7. At one point I had opened multiple embedded flash videos in various tabs, all of which were paused, and Chrome still pegged my cpu at 100%. I'm guessing this was a bug somewhere.
8. I find the little address tab that appears and disappears at the bottom left of the screen when you're loading a new page to be highly distracting. I realize Firefox has a status bar across the bottom of the page, and that it updates when the browser is loading something, but having only the text change is less visibly distracting than having a tab appear and disappear. Especially when it happens rapidly.
Slashdot Mirror
1 NEW
Purists would even disagree with the article author's implied definition of Computer Science, i.e. when she mentions "program design". For some folks "Computer Science" is wholly separate from "programming".
Break up companies such that there are "pipe providers" and "content providers". If my net service provider is not also my cable TV company then it loses much of its motivation to oppose net neutrality. Of course, this may not go far enough. If I'm an ISP and my backbone connection doesn't provide sufficient bandwidth to Comcast's digital cable stream, Comcast will be unhappy (and so will my users). So I'll need to add that bandwidth. The question is how much of that effort should be paid for by me, the "pipe guy" and Comcast (or any other content provider). I'm tempted to say that the market would dictate "pipe provider" behavior without any direct intervention from content providers. For instance, if Comcast offers digital cable and says "our service works on X, Y and Z providers, but not U, V and W", then U/V/W are going to face pressure to properly support Comcast's content stream.
I recently bought several home appliances (washer, dryer, gas oven) at a local appliance store (Conn's). After I'd picked out what I wanted, the store guy went to a computer and looked up web prices on the same items at Best Buy, Lowe's, Sears and Home Depot. Then he knocked about $200 to bring it in line with what one of the other places was offering.
First there's this:
...which is then followed by this:
Here's what confuses me: the first bit seems to suggest the stem cells used to "make the monkey jump" were adult, not embryonic. So why include the last little bit about embryonic stem cell research? Am I incorrect about the first quote, and in fact the cells they used were embryonic in origin?
Define "better".
Point taken. But I also think this approach would change the way people conduct their lives and the way in which local governments operate so as to make both much more expensive and less efficient. If the goal weren't so much "terror" as it was "requiring the addition of security 'overhead' to anything and everything", then this approach would do the trick.
I've always thought terrorists could get a lot more bang for their buck by using much less extravagant means. You don't need to crash a jet into a building; just replicate the D.C. sniper from a few years ago across 50 different U.S. cities. Target local government officials, police officers, women, children, etc. All you need is to get some guys with marksmanship training into the country then get some high-powered rifles into their hands. Instruct them to take their own lives if capture is imminent. I have a feeling this tactic would go a lot further towards "instilling fear" in the U.S. populace on a day-to-day basis than the plane crashing thing.
For such a small organization you might want to keep things as simple as possible given you're not going to have many staff to support 20 employees. Probably "one guy". To that end:
* Branded Gmail for email and calendar. You can use the branded google accounts for IM as well. The spam filtering and uptime are very good. Also you don't have to manage any of it. Your employees can automatically check their work email from anywhere w/o having to get on a VPN or use a particular email client.
* Macs. Makes you functionally immune to malware. Repairs/replacement are pretty speedy, esp. if you have a Mac store in your area. If you absolutely positively must have MS Office then you can get it as a native app. If you must run Windows then there's a free virtualization option (Virtual Box). It's not as good as Parallels, but it gets the job done.
* If you expect your employees to occasionally work from home (or on the road) or if you want to at least give them that option, then get everyone laptops. Providing external monitors, keyboard, mouse is fairly cheap. MacBooks (not MacBook Pro) are in the same ballpark cost-wise as similarly spec'd name-brand PC laptops.
* Hosting your own web server sounds like an unnecessary pain in the butt. If you absolutely must, then Linux/Apache is probably the way to go. I'd recommend the latest LTS version of Ubuntu (10.04). Going with an esoteric distribution just makes finding documentation and fixing problems that much more time-consuming.
Use it until they make you pay for it or the features become unbearably annoying. Who cares if its from Oracle? Its not like you're spending money on it. Personally I've always found OO to be a giant pile of crap and use MS Word instead. 2003 version; I don't especially like the new 2007 look and feel.
Define "drying up". If you look at the percentage of non-Microsoft web-based development positions based around Java and PHP...its a pretty high percentage. Personally I wouldn't waste my time learning a niche language/framework like RoR. Learning to do Android or iPhone development would probably be useful.
For maximum simplicity just move everyone to OS X. The guys who need Office can use MS Office for the Mac. The non-technical users won't be freaked out by OS X in the way they might be by Linux. OS X gives you most of the same malware immunity you get with any other non-Windows OS. The marketing and graphics guys get to keep using Macs just like they always have. Your developers, if there are any, should be fine on the Mac unless they're doing development that specifically targets Windows.
From a cost perspective, of course, this may not be the cheapest thing in the world. But you can't beat it for simplicity. I also like the branded Gmail suggestion. That can work irrespective of which OS's you deploy.
I think you misunderstand me; generally speaking like OS X. Furthermore, I suspect the fact that it lacks some of the irritations of, say, an Unbuntu, is largely because of Apple's dictatorial control and closed environment. OS X seems to hit a nice sweet spot between non-technical usability and not completely shafting technical users. I appreciate being able to drop to a bash prompt and do unixy things when I need to. I like not ever being forced to do that.
This article seems like a big excuse. If you can't beat 'em then don't try. Want a great example of a Unix-like OS beating MS at their own game? OS X. Know what it took? Some quality building blocks, then dictatorial stewardship of a single company with deep pockets and a willingness to meet the needs of a completely non-technical user.
I disagree with the "idiocracy" tag on this article. It is perfectly rational behavior for me to push the "walk" button at a crosswalk even if there's a decently high chance it will have no effect. Why? Because the cost to me is very small vs. a potentially large payoff if it actually does work.
I love that Phoronix is willing to take the time to run tests like this. I just wish they'd learn how to run meaningful tests. For instance, why are they testing a bunch of CPU-bound things? Kernel won't affect that unless we're talking about SMP performance. If you want to test the kernel, test how well it handles SMP, network I/O and disk I/O. And bear in mind that disk I/O will be hugely affected by which filesystem is used and its configurable settings.
Another problem with their article is that it tests individual kernels. Most folks don't use a vanilla kernel. They use one provided by their distro, which may have distro-specific patches that address some of the performance problems (or add new ones). What I would have preferred to see is a comparison of different distro releases over the last 5 years, focusing on the most popular ones (say Ubuntu, Fedora and SuSE).
The meaningful tests (and their results) were:
1. GnuPG: avoid 2.6.30 and later.
2. Loopback TCP: avoid 2.6.30 and later.
3. Apache Compilation: avoid 2.6.29 and earlier.
4. Apache static content: avoid 2.6.12, 2.6.25, 2.6.26, then 2.6.30 and later.
5. PostMark: avoid 2.6.29 and earlier.
6. FS-Mark: avoid 2.6.17 and earlier, 2.6.29, then 2.6.33 to 2.6.36.
7. ioZone: unless you're willing to run 2.6.21 or earlier, avoid 2.6.29 and you're fine.
8. Threaded I/O: avoid 2.6.20 and earlier, 2.6.29, then 2.6.33 to 2.6.36.
Based on these results, #1 and #2 seem to be testing the same thing, and tests #3 and #5 seem to be testing the inverse of whatever that thing is. 2.6.29 seems to be especially crappy, performing worse than the kernels immediately before and immediately after it on tests #6, #7 and #8. In terms of recent kernels, tests #6 and #8 suggest a regression in 2.6.33 that has been resolved in 2.6.37.
If it were me, I'd look at either running 2.6.37 (when its released) or fall back to 2.6.32 if my hardware was supported.
Yes. I've long maintained that a course in prob/stat would be way more useful than, say, trigonometry and calculus for students who aren't going into science/engineering professions. Honestly if you could just ram home two basic concepts: 1. correlation is not causation, and 2. what "statistical significance" means, then that would serve most folks well.
I was going to post this as well, but I'll just agree with what you wrote. Originally Apple probably rolled its own JRE because it didn't think it had a large enough Java user base to force Sun to target OS X. Apparently Apple now thinks it does have a large enough user base to force Oracle's hand. So they stop rolling their own and let Oracle take the heat if it decides to abandon Mac users.
Though I'd say it was always pretty much dead. Linux on the desktop is fine for, say, a developer workstation. Or maybe something like a netbook or smartphone OS where the content is mostly managed and/or runs in a sandbox that's entirely shielded from the linux environment (e.g. Android). Anything else its less trouble to run Windows or OSX.
Define "need". Then define exactly which (and how many) humans. Then we can talk.
Exactly. Here are Microsoft's statistics after computer prevalence is taken into account. Quote from that page:
Given the very low infection rate of most of Africa, though, something tells me Microsoft's "CCM" metric may not perfectly reflect real infection rates.
I worked for a startup that used branded google services for email, instant messaging and calendar. It worked pretty well. Not sure how the cost worked out, but it was certainly less headache than maintaining our own mail server. Especially since we didn't have a dedicated IT person.
Whine much?
Here are my complaints in no particular order:
1. It won't open PDFs in a browser window.
2. While its sync functionality does run automatically at browser shutdown, it currently doesn't sync stored passwords.
3. Since there is no Application Bar in XP like there is on OSX I have to configure Chrome to have the "bookmark bar" visible on every tab.
4. On XP Chrome exhibited a UI bug where the horizontal space at the bottom of the screen used to display downloaded files became permanently "blank". I wasn't able tor reproduce it, but the fact that it happened at all was annoying.
5. Scrolling on XP was slow and jumpy compared to Firefox and IE. Not sure what's going on there. I tested it side-by-side w/ Firefox so I know it's not my mouse or mouse driver.
6. While Chrome is faster on JS-intensive tasks it "feels" slower on regular page loads. My suspicion is that this is due to Chrome starting to render the page before all elements are received, which causes the page to visibly change as new elements are loaded. Firefox has a short delay where there are no visible changes then displays the page all at once. For whatever reason I find this more pleasing.
7. At one point I had opened multiple embedded flash videos in various tabs, all of which were paused, and Chrome still pegged my cpu at 100%. I'm guessing this was a bug somewhere.
8. I find the little address tab that appears and disappears at the bottom left of the screen when you're loading a new page to be highly distracting. I realize Firefox has a status bar across the bottom of the page, and that it updates when the browser is loading something, but having only the text change is less visibly distracting than having a tab appear and disappear. Especially when it happens rapidly.
Yes, I could. I just got through trying Chrome for a week. No thanks. I might give Opera a try if Xmarks goes away or becomes unacceptably expensive.