You allow others to edit your CV? I send my resume out as a PDF specifically to make sure that they don't edit it. Too many problems with people thinking it okay to mangle my resume, redoing the formatting, in one case even sending raw HTML garbage because they didn't even know how to forward a HTML document on.
Upgradability hides the real advantage I have of a desktop, the ability to affordably repair it. The last several purchases I've made for my computer were to replace broken components that are integrated in laptops. Speakers (admittedly easy to plug in a new one on a laptop), keyboard, and monitor were three of those upgrades made in the past few years.
Compare this to the people I know with laptops. Several of them had to replace the entire unit when things broke like the sound card, keyboard, screen, etc.
My wife's system is similar. Keyboard replacement, external mouse replacement, things like that. Very expensive and annoying things on a laptop. Easy on a desktop.
Re:Not going to happen, ever _ minus Centagrade
on
New Calendar Proposal
·
· Score: 2, Insightful
Farenheit is a more convienient method precisely because the commonly used temperatures for everyday usage are between 0 and 100. In the centigrade scale, at least half the scale is not typically used.
People make a lot of noise about how "superior" the metric system and I simply sit back and laugh. I see the whines about not understanding ounces and pounds and then these same people go on to talk about using hexidecimal numbers as routine. (In case you didn't realize, there are 16 ounces in a pound, 16 fluid ounces in a pint, and "a pint's a pound the world 'round").
The metric system hasn't won out precisely because it isn't inherently "superior" in any way. I suspect that the whining over the English system is just a meme that dates back to some mathematically illiterate folks who thought that the only way to handle anything was to make it base ten.
For grade school material, I can handle the teaching myself, the issue is the school that actually discourages teaching of standard arithmetic and declare the child who actually understands the algorithm as only have a fraction of the understanding of the math simply because they never bothered looking at the problem in the way the teachers want.
So the problem arose completely because I offered supplemental material at home. By the time he's in middle school, one of two things is likely. 1. He will be ruined on math because he will have grown sick of teachers who make him repeat the same boring stuff he learned months upon months ago. 2. He will have such a self superiority complex that he will have a very hard time when he does actually get challenged in math because he won't know how to study, how to actually work at learning it anymore because no teacher ever bothered giving him a challenge before.
Tell me about lazy kids and schools not being accountable. I just came today from a meeting with my son's school teachers about his math program. My son wants to do more advanced math work, has been ready for it for some time. The school's response was to claim that the ability to perform arithmetic on paper has little to do with mathematics and then deny him access to ability appropriate mathematics. So a child who has been doing full multi-digit addition and subtraction with carrying and borrowing is asked to do single digit addition with answers no higher than 15 as the most advanced math they will offer him.
Part of it is the National Council of Teachers of Mathematics (NCTM) has allegedly endorsed a program that deemphasizes pencil and paper arithmetic to the point that some of the more extreme advocates of this program have proposed banning traditional algorithmic arithmetic until close to fourth grade.
The listed criteria that the school has identified as necessary skills are available at the NCTM website.
This list may look initially acceptable, but the application of it at least in my son's school was to claim that arithmetic is not even a significant part of math, at least not a standard algorithmic understanding of how to do the standardized problems. Instead, an emphasis on "strategies" is supreme to the point that if a problem cannot be done in one's head, it isn't worth doing.
The other issue is the "No Child Gets Ahead Act". It requires teachers to bring up to minimal standard as many students as possible and ignore those students who meet the minimum requirements without trying. This approach discourages advanced work in all too many cases that I have seen.
There are often problems with lazy students, but that is not the whole of the situation, overly rigorous school programs are just as much to blame.
Find me three pieces that would be playable by less than a college level performer that use a 5/4 time signature. Much more common would be 2/4 which is often used in music a child of that age might sing or perform. (After just helping my 7 year old prepare for a recital with a piece in a 2/4 time signature,...)
As for the chores, those sound typical for any child of that age. But forget the fresh basil, get the fresh catnip instead. Use it as mint in your cooking and then rub the extra on whatever you want the cats to go crazy over. I find that getting good half sharp paprika (not that tasteless garbage you find in most grocers) is far more useful to me than basil.
But why would you want a child to use that bastardized obsolete handwriting system called cursive? I haven't used it for more than my signature (which is rapidly becoming less of a cursive over the past two years) since fourth grade. It is very difficult to read by a human, never mind a computer, and is really not enough faster than print to justify the cost in reading it. If I want to prepare something quickly, I type it.
I talked to a vendor once about a tankless water heater. After asking a few questions, he told me he would not install one for anyone in my neighborhood. The drastic question? "How well do you deal with water with a high total dissolved solids (TDS) level?" The installer who was pushing them as the ultimate solution was saying that he would ruin my pipes. One reason water heaters go every so many years is sediment and gunk from impure water builds up on the bottom.
I've had to pick up the pieces before of a homeschooled child. In the case in question, the child started out as on grade level in math, two years above grade level in English, good at one foreign language (French), etc. All in all, a moderate to good student. After one year of home schooling, the English had deteriorated to grade level, the math level had fallen to more than a year below grade level, and all other areas had similarly atrophied. This student was not plopped in front of a TV, in fact, the parent was a former English teacher (retired). What happened? People forget that the role of the teacher and role of the parent are not the same thing and should at best rarely be mixed.
Vouchers are bad because they take more money from the local school system than the real cost of the student to the school system. If one has a class of 24 students, then 3 of those students leave on a voucher, the actual decreased costs to the school are fairly minimal, but the amount of money taken from the school to educate the remaining 21 students is not. Vouchers are also based on the notion that parents can choose not to contribute to program. It is a dangerous concept. Do we start listing other programs like roads and if a person decides that they don't want to use the services of that program, they can get a voucher to pay for an alternative? Think of police, fire, or emergency medical services as an example. Even less urgent programs like the way pay phones are subsidized in some communities to ensure that they remain in some areas where they would otherwise be removed for unprofitability.
Vouchers are also based on the premise that the students perform better in the private schools. Dr. Witte's analysis did not support that conclusion. His analysis of the Milwaukee voucher program (one of the oldest and best studied programs in the nation, the Cleveland program may be larger though) did not find statistically significant improvement in academic performance of voucher students over their peers who remained in the public schools, this over a ten year period. In fact, the primary advantages he cited were in the area of parental satisfaction and access to extracirricular programs. (Note, the Milwaukee voucher program was targetted at low income families who could not easily afford private schooling.) Unfortunately, fraud became an issue with the program.
Uniforms are fairly common today at public schools. Not to mention the fact that not all private schools have uniforms, so that can't be it. (Find me a public school that doesn't have a no-hat uniform code today. Some schools even prohibit hats while the student is outside.)
Many private schools are secular, so that can't be it. (Here I always thought religious instruction was the job of the parent.)
One thing that a lot of people forget is that private school teachers are not paid as well as public school teachers as a rule. It is precisely because private school teachers are poorly paid that the cost per student can be so low. When the cost per student of K-12 approaches the cost per student of out of state tuition at a four year college for a similar number of classroom hours, then I'm interested. K-12 has always seemed to me to be more expensive than college education in terms of cost per student, not less, yet we expect to do better with less money per student, even if the teachers are paid only moderate wages.
Face time with your colleagues isn't as essential or common as one might think if an effort is made.
On my current project (I've been on it about 4 years now), I've never even met my boss of around 17 months, only met my last boss twice when we both went to different meetings in the same city at the same time, and never have met most of my coworkers. We aren't even all teleworkers (just me in fact). Working remotely has many of the same problems of teleworking. What difference does it make where I am working if part of my team is in California, part in the MidWest, part in Texas, and part in New York?
That said, contact with colleagues can be maintained, but it takes training and a consistent effort. It means literally calling up your peers and talking with them on a semi-informal basis much like walking over to their cube and chatting. It also takes an effort on everyone's part to communicate better in email or similar methods. Regular conference calls are a must to ensure that everyone is on the same page on the project. There are business training courses available to help train people in teleworking, what preconditions must exist to make it a success and how to cope with it. How to make it work is similar to how to make a remote office situation work where your project team is all over the country.
One of the basic rules of outsourcing is you don't outsource your mission critical work. If you do, then why does the company exist at all instead of the outsource firm doing it directly without the overhead of the other company?
Also, some jobs are just fundamentally a bad idea to outsource because of the issues with continuity and corporate security. Examples of this include your internal corporate security department.
There is also little difference between teleworking from a different office and teleworking from home. As someone who has telecommuted for the past seven years, I started not because of some proclaimed convienience factor, but because my official office had no one I worked with in the same building. A couple years later, I didn't work with anyone within a few hundred miles. Yet being on the corporate network and a corporate employee (instead of an outsourced contractor) makes my job far easier for me. Our outsource sites are constantly fighting a lot of issues of network access, management structure, etc. that I just don't have to deal with.
Seriously. I have found this entire ATM craze to be completely stupid. If I want cash, I go to the bank (which is located inside most of the major grocery stores in the area, and yes, I do mean inside) which has almost no lines unless I hit it on a Saturday morning right before they close. There are no fees for writing a check made out to cash at any bank I've ever checked at. It's just standard operating procedure. You have written a valid check, there shouldn't be any fees on that unless you are some dumb idiot who got an account that has a fee on every check you write. Then you simply are taking a check written out to you and cashing it, again a transaction that should be no fee.
Any bank that wants to charge fees, you just tell them that you will take your money elsewhere unless they change their policy.
Some of the grocery stores even allow you to write a check for up to $20 over the amount of the purchase so you can get some cash back when you buy your groceries if for some reason you can't hit the grocery store while the banks are open.
I don't even use checking withdrawal slips anymore, I just write a check for the appropriate amount.
As for convienience factor? I've found checks accepted almost everywhere I go as long as I stay in state except for restaurants, and at those places, I tend to use credit or cash anyway. (And now some restaurants take check).
The only down side is some businesses have stopped taking check and instead do a "Electronic Check Conversion" which is you authorizing the merchant to make an arbitrary sized withdrawal from your account. There is no legal fraud protection on those, so if they hand you back your check and a little credit card receipt to sign, don't do it. Announce loudly that you don't agree with such, and that you are doing a check transaction, not a electronic debit transaction. They are in theory required to notify you in advance of the transaction for any mail order transactions along with a few other regulations.
I don't even have an ATM card, never needed one. The bank should pay me for the priviledge of holding my money, not the other way around.
I've seen someone threatened with arrest for asking for the badge number of a police officer. Just because most good officers have no qualms providing badge number/ID to prove their identity doesn't mean that others won't respond with threats.
I have yet to find an area where even half of the school board members held the level of education of any of the academic subject high school teachers. As a typical example, one person I knew would regularly challenge all prospective school board members to see if they had a basic understanding of first semester calculus (which was taught to almost certainly more than 200 students per year in the school district). It was rare to find more than one canidate with that level of understanding, despite that being considered the first non-remedial math course at many colleges. It is important to realize that I am not talking about rural areas with very few members of the community who have college degrees, never mind advanced degrees. These were decent sized communities with a good sized technical workforce.
You ask who would serve on a school board if they didn't truly value education? Did you see the incidents in Kansas and the numerous other incidents at local school boards where people who clearly value a lack of education go for the school board to ensure that the children are not educated? I've seen people running for school board (and elected) who openly opposed public education as a concept, wanting to disband it. I don't think I was watching just a strange case.
Which brings up the real reason this list is worthless. We as Joe Q Public have little to no way of knowing what kind of supercomputers were delivered in secret or confidentially.
It's a brag fest for those who buy supercomputers and don't care who knows their real compute power. I have a gut feeling that is actually a rather small set of people. If you read the About, they are based on surveys of manufacturers and users.
Another note of interest is that the benchmark chosen is to solve a dense system of linear equations. Depending on what kind of problem you are buying supercomputer time for, that may not be the best benchmark. What if you are trying to solve problems based on very large sparse matrices and very fast random memory access is the defining bottleneck instead? Or another problem may require one to use integer math instead of floating point math? Such things can radically change the performance of a system. (Or to be really evil, what if your problem does not parallelize well at all? Suddenly a cluster may not be anywhere near as useful as it once was.)
As someone who has introduced a kid to both Windows and Linux on a dual boot system (he is now 5), the underlying OS didn't matter one bit. He only cared "can I play my favorite programs?". Wasn't overly difficult to convince him to slowly migrate over to gcompris. It certainly helped that gcompris has been actively developed recently.
For a young child (under age 5), a text processor like emacs or kwrite, some educational software (gcompris), and a painting program (there's one in gcompris, or use the one that comes with KDE, I have both KDE and GNOME on the system) seems to help keep my kid interested plenty. For games, he's so addicted to tux racer that I doubt I could move him off that game if I tried. (Yes, this is the commercial version.)
Yes, kids like to hit random buttons. Way back when, my kid would teach me some interesting built in macros of OS/2. Now, I set him up on a carefully constructed KDE desktop with very few icons on it (double clicking used to be a problem for him), and he is a lot less dangerous.
The first way is for people to be straight with their employer that they follow the SAGE code of ethics. The best time to do that is before one is even hired (for example, on one's resume). If one is already hired, then mentioning it to ones manager in a polite and nonconfrontational manner is probably a good idea. It helps if one has already made clear that one is a member of SAGE.
Another method that works well is printing it out and posting it on your cube wall so people walking by your cube can see it.
Another method that can work is by becoming a certified SAGE SA (see http://www.sagecert.org/)
. One of the requirements of the cSAGE certification is to have read the code of ethics. Note, that does not mean one has to follow it, but one does need to have read it. The more cSAGE certified SAs on your team, the more your management will be aware of the certification, and the more reasonable the idea of a code of ethics will become.
As a senior SA in the team I work for, I made it clear to all the other people in my team that I expect them to be aware of the SAGE code of ethics. Again, I am not saying that they have to follow it, because I am not their supervisor. But it raises the awareness of the code of ethics.
As for making people more aware of SAGE itself, one excellent method is by showing up at SAGE
local groups that are in your area. Make sure
your manager knows what you are doing. Even if
you are on call, frequently a deal can be worked
out that someone else will cover for you while
you are attending that meeting. Make sure to
let your manager know that it is more than a social club, discuss some of the ideas that come up in the meetings. That way, your manager gets some constant reminders about SAGE itself, and some of the advantages that exist for it.
If you want good SAs, go look at SAGE. Or advertise for a job offering, insisting that your SAs be cSAGE certified. Don't whine about SAs, hire a good SA and make them the team lead. A good team lead SA has the authority to make things done correctly. You get the advantage of at least one good SA to do a brain transfer, the possibility that the other SAs will improve, and the advantage that the other SAs have a hard time telling the lead SA off if they want a good reference at their next job.
The other advantage of a good senior SA is to teach religion. The religion that menus belong in restaurants, and the function of a mouse is to do the nifty xterm select/paste with the middle mouse button. Contractor SAs tend to dislike automation (horrible, but true), because automation means less money in their pocket.
As for how to deal with such an environment, it does take some setup work, and a different way of thinking. I use cfengine heavily to make my life livable.
Careful work with vendors to ensure that fingerpointing won't occur is a must. Some vendors are obviously better at that than others. Usually I insist that regardless of who is at fault, every vendor involved will make appropriate experts available from the time the problem is detected until the time the problem is certified to be resolved (even if that means that the support staff hands off three or four times because their shift has ended).
Don't be surprised that SAs get annoyed with applications messing up servers. I've seen more than one case where an application was so badly coded that it brought the server to its knees. A few minor changes to the application to make it less of a brain dead design, it was humming along quite nicely without any significant impact to the server. I know I'm not the only SA to have seen things like that. I'm not saying that such is always the case, but it does happen. Maybe, just maybe, an SA knows what they are talking about when they say the application is the cause of the problem.
Where I work, I've seen the entire gamut. I've seen SAs who were great at scripting, but didn't understand simple computer security. I've seen SAs who while they weren't great scripters, I could hand them almost any troublesome problem and they would keep at it until they came up with a real solution, even if it took them a while. SAs aren't so much declining in quality as I see more and more need for SAs, and the burn out rate has gone up. Fewer and fewer old dinosaur SAs are left as they burn out and leave, and more junior SAs are too likely to burn out and leave SA before they ever become a senior SA. (It wasn't that long ago that it was common for servers to be able to be down every night for a few hours, now, it's rare that I get a server that is allowed a few hours a week of downtime.)
Back in 1991, I saw a high school student attempting to replicate a published result that looked very similar to the description in the article. The device was different, but the concept of very small weight loss of a spinning object was the same. In this variant, a high speed gyroscope was used. Turned out, the key was the electricity going through the wires to power the gyroscope and make it spin. The high schooler showed that if the electric current (going through wires) came from different perspectives, the "weight loss" could appear or disappear.
While I haven't read the extensive details, I do have to wonder if we might be seeing a similar concept here. What made me especially suspicious is the line that the effect only appears at several thousand RPMs. Turning up those RPMs means more current to the system, and more potential for side effects. Since the primary effect being looked for is very subtle, it could become significant.
One place I see a need for this kind of behavior is the user who knows enough to be interested in running linux, but isn't comfortable with programming and thus is not comfortable downloading and compiling a new kernel. The distro kernel upgrades are usually nice, but in one particular case I saw, (Mandrake 8.1), using the distro supplied tool to update the system kernel (from the distro kernel, not even a recompiled kernel from distro source) made the system unusable. I had to step in and grab a newer kernel and then compile it.
Even a reasonably seasoned user who knows what they are doing could take advantage of at least an autoconfigure type kernel utility that probed your hardware and filled in entries appropriately. For example, why should my kernel try to compile in code to fix a bug in chipset xyz if I don't have that chipset and another chipset was correctly identified in its place? It could save the trouble of looking at various outputs of lspci and similar to try to figure out exactly what hardware is on the system you are working on, something that many users may have difficulty determining.
While many of these ideas sound great in theory, the reality of administration forces many of these ideas to be abandoned.
1. Physical security. This should already be in place, and not just for credit card numbers. If you don't have adequate physical security on your servers, you have a tremendous problem. Most decent hardware has the ability to key lock access to the hard drives.
2. Network operations monitoring. Often overlooked, this feature is crucial, and contradicts some of the features others advocate of eliminating all other network connections, etc. Any service that needs to be up all the time, such as a billing server (needs to receive new records all the time for changes to billing, as well as customers querying the status of their bill, etc.), should be monitored by a dedicated operations group. The monitoring should be mostly automated (Big Brother comes to mind as a free utility, there are others both free and commercial), but with a human to evaluate the alerts you get so your admins last more than 3 months. This kind of monitoring nicely encourages a smart person to throw in some log monitoring to monitor the logs for "unusual" network activity to flag an alert. But you obviously can't yank the network cable every time you receive an extraneous port 21 connect. You need a human around to look for patterns and have the brains to know when to wake the admin and when not to. You need a dedicated staff of at _least_ 3 people (one off call, one primary on call, one secondary on call) at all times.
3. What I have found works well is to segment the process into multiple servers. Thus you can have a web server that can submit an append request to the database server, but can't do a full read, only certain types of read requests. The database server would of course be sequestered as it would hold your real data.
4. Forget locking people down to the console only. That works up until you need three people working at once to resolve an outage situation. (Don't say that doesn't happen). Instead, use an isolated network with a "hop" box that allows you to choke point access. Not the same as a firewall, but a firewall could serve the purpose as well.
5. As many others have said, make sure all traffic in and out of the box is encrypted by some method. PGP works well for some things, SSH works well for others. There are other options out there. (If you are talking to mainframes, there are some other interesting options.)
6. Good backups are essential. Sooner or later, some idiot will remove a key file at 6PM Friday, only to be noticed at 6AM Monday (or worse, not until the first of the month when you do a large monthly billing run.) There are commercial services out there that will securely store your backup tapes offsite for a given retention period. Test your backups also. On your test machine, blow away the database (prior to going live preferrably), and then restore the system from your backup tapes, and do that at least once every other year, restoring random files at least 3-4 times a year. What good is your data if it is unavailable when you need it?
7. Test machines are a must. When you want to test to see if you left open a back door, don't do it to your production system, do it to your test system that is an exact duplicate in terms of hardware type, every piece of software, and network layout (but obviously no customer data, just dummy data that emulates the structure of the database). It does no good to secure a system only to have the security compromised because you didn't anticipate a side effect that actually chokes off your primary application's access.
8. I don't know if the vendor selected uses phone validation, but if that is the route you pick, make sure your lines are configured for dial out only. Otherwise, it can leave an interesting hole that won't be revealed by even a very heavy TCP scan or lsof -i TCP
First of all, Sun CPUs are far from outstanding. Until the UltraSPARC III came out (which many people do not have yet), they were much slower than most of their workstation/server competitors. (I am assuming you are talking about workstations here because no one should be stupid enough to have a desktop environment running on a server.)
Second of all, remember that if your system is doing nothing, it is very generous to processes that would otherwise do nothing. From what you say, it sounds like some kind of network access or other runaway process is taking up your CPU time. Check your mpstat, your netstat -i, etc. to look at your bottlenecks. You could have some combination problem going on that causes a performance problem when two things are running at the same time.
As for comparison of HP's VUE (excuse me, CDE,) to GNOME, a lot of that is personal preference and habit. I used HP VUE prior to CDE coming out, and then switched to CDE and found the differences relatively minimal, even in the config files. I use GNOME at home, my wife uses KDE. My major complaint about GNOME is that I prefer the old xterm to ANY of the newer terminals I've used. (Get rid of those stupid menus, and give me back the xterm scroll bar, complete with the "strange" mouse control.) CDE's pager is mildly annoying to me, but that's because the first pager I used was similar in style to the one in FVWM and GNOME rather than CDE.
CDE is dead because of a squabble between the major CDE "partners". Ever try to get HP's CDE and Sun's version to cooperate? It's effectively impossible. The two sides are squabbling so much that CDE has not fundamentally changed since 1996 so far as I can see. It is clear that something needs to be offered soon to desktop users soon with proper commercial support. Maybe GNOME isn't the right answer yet with nautilus. (I prefer gmc myself for what little I use a file manager for.) But something does need to be done.
There is a major reason for the RC5 effort that you seem to be overlooking. Prior to the 24 hour DES crack, it was illegal to use DES in France. Shortly afterwards, France decided that DES was legal. I don't think that was coincidental. The RC5-64 crack effort is partly an effort to encourage governments to permit broader use of strong cryptography, or in the case of the United States, export stronger cryptography.
Cracking challenges also encourage people to look heavily at implementations of cryptographic algorithms to see if they can find shortcuts. That is research. It may be unlikely to result in much return, but there has been more than one example of a cryptographic implementation that had problems.
Slashdot Mirror
You allow others to edit your CV? I send my resume out as a PDF specifically to make sure that they don't edit it. Too many problems with people thinking it okay to mangle my resume, redoing the formatting, in one case even sending raw HTML garbage because they didn't even know how to forward a HTML document on.
The lack of easy editability is bonus for me.
Upgradability hides the real advantage I have of a desktop, the ability to affordably repair it. The last several purchases I've made for my computer were to replace broken components that are integrated in laptops. Speakers (admittedly easy to plug in a new one on a laptop), keyboard, and monitor were three of those upgrades made in the past few years.
Compare this to the people I know with laptops. Several of them had to replace the entire unit when things broke like the sound card, keyboard, screen, etc.
My wife's system is similar. Keyboard replacement, external mouse replacement, things like that. Very expensive and annoying things on a laptop. Easy on a desktop.
Farenheit is a more convienient method precisely because the commonly used temperatures for everyday usage are between 0 and 100. In the centigrade scale, at least half the scale is not typically used.
People make a lot of noise about how "superior" the metric system and I simply sit back and laugh. I see the whines about not understanding ounces and pounds and then these same people go on to talk about using hexidecimal numbers as routine. (In case you didn't realize, there are 16 ounces in a pound, 16 fluid ounces in a pint, and "a pint's a pound the world 'round").
The metric system hasn't won out precisely because it isn't inherently "superior" in any way. I suspect that the whining over the English system is just a meme that dates back to some mathematically illiterate folks who thought that the only way to handle anything was to make it base ten.
For grade school material, I can handle the teaching myself, the issue is the school that actually discourages teaching of standard arithmetic and declare the child who actually understands the algorithm as only have a fraction of the understanding of the math simply because they never bothered looking at the problem in the way the teachers want.
So the problem arose completely because I offered supplemental material at home. By the time he's in middle school, one of two things is likely. 1. He will be ruined on math because he will have grown sick of teachers who make him repeat the same boring stuff he learned months upon months ago. 2. He will have such a self superiority complex that he will have a very hard time when he does actually get challenged in math because he won't know how to study, how to actually work at learning it anymore because no teacher ever bothered giving him a challenge before.
Tell me about lazy kids and schools not being accountable. I just came today from a meeting with my son's school teachers about his math program. My son wants to do more advanced math work, has been ready for it for some time. The school's response was to claim that the ability to perform arithmetic on paper has little to do with mathematics and then deny him access to ability appropriate mathematics. So a child who has been doing full multi-digit addition and subtraction with carrying and borrowing is asked to do single digit addition with answers no higher than 15 as the most advanced math they will offer him.
Part of it is the National Council of Teachers of Mathematics (NCTM) has allegedly endorsed a program that deemphasizes pencil and paper arithmetic to the point that some of the more extreme advocates of this program have proposed banning traditional algorithmic arithmetic until close to fourth grade.
The listed criteria that the school has identified as necessary skills are available at the NCTM website.
This list may look initially acceptable, but the application of it at least in my son's school was to claim that arithmetic is not even a significant part of math, at least not a standard algorithmic understanding of how to do the standardized problems. Instead, an emphasis on "strategies" is supreme to the point that if a problem cannot be done in one's head, it isn't worth doing.
The other issue is the "No Child Gets Ahead Act". It requires teachers to bring up to minimal standard as many students as possible and ignore those students who meet the minimum requirements without trying. This approach discourages advanced work in all too many cases that I have seen.
There are often problems with lazy students, but that is not the whole of the situation, overly rigorous school programs are just as much to blame.
Find me three pieces that would be playable by less than a college level performer that use a 5/4 time signature. Much more common would be 2/4 which is often used in music a child of that age might sing or perform. (After just helping my 7 year old prepare for a recital with a piece in a 2/4 time signature,...)
As for the chores, those sound typical for any child of that age. But forget the fresh basil, get the fresh catnip instead. Use it as mint in your cooking and then rub the extra on whatever you want the cats to go crazy over. I find that getting good half sharp paprika (not that tasteless garbage you find in most grocers) is far more useful to me than basil.
But why would you want a child to use that bastardized obsolete handwriting system called cursive? I haven't used it for more than my signature (which is rapidly becoming less of a cursive over the past two years) since fourth grade. It is very difficult to read by a human, never mind a computer, and is really not enough faster than print to justify the cost in reading it. If I want to prepare something quickly, I type it.
I talked to a vendor once about a tankless water heater. After asking a few questions, he told me he would not install one for anyone in my neighborhood. The drastic question? "How well do you deal with water with a high total dissolved solids (TDS) level?" The installer who was pushing them as the ultimate solution was saying that he would ruin my pipes. One reason water heaters go every so many years is sediment and gunk from impure water builds up on the bottom.
I've had to pick up the pieces before of a homeschooled child. In the case in question, the child started out as on grade level in math, two years above grade level in English, good at one foreign language (French), etc. All in all, a moderate to good student. After one year of home schooling, the English had deteriorated to grade level, the math level had fallen to more than a year below grade level, and all other areas had similarly atrophied. This student was not plopped in front of a TV, in fact, the parent was a former English teacher (retired). What happened? People forget that the role of the teacher and role of the parent are not the same thing and should at best rarely be mixed.
Vouchers are bad because they take more money from the local school system than the real cost of the student to the school system. If one has a class of 24 students, then 3 of those students leave on a voucher, the actual decreased costs to the school are fairly minimal, but the amount of money taken from the school to educate the remaining 21 students is not. Vouchers are also based on the notion that parents can choose not to contribute to program. It is a dangerous concept. Do we start listing other programs like roads and if a person decides that they don't want to use the services of that program, they can get a voucher to pay for an alternative? Think of police, fire, or emergency medical services as an example. Even less urgent programs like the way pay phones are subsidized in some communities to ensure that they remain in some areas where they would otherwise be removed for unprofitability.
Vouchers are also based on the premise that the students perform better in the private schools. Dr. Witte's analysis did not support that conclusion. His analysis of the Milwaukee voucher program (one of the oldest and best studied programs in the nation, the Cleveland program may be larger though) did not find statistically significant improvement in academic performance of voucher students over their peers who remained in the public schools, this over a ten year period. In fact, the primary advantages he cited were in the area of parental satisfaction and access to extracirricular programs. (Note, the Milwaukee voucher program was targetted at low income families who could not easily afford private schooling.) Unfortunately, fraud became an issue with the program.
Uniforms are fairly common today at public schools. Not to mention the fact that not all private schools have uniforms, so that can't be it. (Find me a public school that doesn't have a no-hat uniform code today. Some schools even prohibit hats while the student is outside.)
Many private schools are secular, so that can't be it. (Here I always thought religious instruction was the job of the parent.)
One thing that a lot of people forget is that private school teachers are not paid as well as public school teachers as a rule. It is precisely because private school teachers are poorly paid that the cost per student can be so low. When the cost per student of K-12 approaches the cost per student of out of state tuition at a four year college for a similar number of classroom hours, then I'm interested. K-12 has always seemed to me to be more expensive than college education in terms of cost per student, not less, yet we expect to do better with less money per student, even if the teachers are paid only moderate wages.
Face time with your colleagues isn't as essential or common as one might think if an effort is made.
On my current project (I've been on it about 4 years now), I've never even met my boss of around 17 months, only met my last boss twice when we both went to different meetings in the same city at the same time, and never have met most of my coworkers. We aren't even all teleworkers (just me in fact). Working remotely has many of the same problems of teleworking. What difference does it make where I am working if part of my team is in California, part in the MidWest, part in Texas, and part in New York?
That said, contact with colleagues can be maintained, but it takes training and a consistent effort. It means literally calling up your peers and talking with them on a semi-informal basis much like walking over to their cube and chatting. It also takes an effort on everyone's part to communicate better in email or similar methods. Regular conference calls are a must to ensure that everyone is on the same page on the project. There are business training courses available to help train people in teleworking, what preconditions must exist to make it a success and how to cope with it. How to make it work is similar to how to make a remote office situation work where your project team is all over the country.
One of the basic rules of outsourcing is you don't outsource your mission critical work. If you do, then why does the company exist at all instead of the outsource firm doing it directly without the overhead of the other company?
Also, some jobs are just fundamentally a bad idea to outsource because of the issues with continuity and corporate security. Examples of this include your internal corporate security department.
There is also little difference between teleworking from a different office and teleworking from home. As someone who has telecommuted for the past seven years, I started not because of some proclaimed convienience factor, but because my official office had no one I worked with in the same building. A couple years later, I didn't work with anyone within a few hundred miles. Yet being on the corporate network and a corporate employee (instead of an outsourced contractor) makes my job far easier for me. Our outsource sites are constantly fighting a lot of issues of network access, management structure, etc. that I just don't have to deal with.
Change banks.
Seriously. I have found this entire ATM craze to be completely stupid. If I want cash, I go to the bank (which is located inside most of the major grocery stores in the area, and yes, I do mean inside) which has almost no lines unless I hit it on a Saturday morning right before they close. There are no fees for writing a check made out to cash at any bank I've ever checked at. It's just standard operating procedure. You have written a valid check, there shouldn't be any fees on that unless you are some dumb idiot who got an account that has a fee on every check you write. Then you simply are taking a check written out to you and cashing it, again a transaction that should be no fee.
Any bank that wants to charge fees, you just tell them that you will take your money elsewhere unless they change their policy.
Some of the grocery stores even allow you to write a check for up to $20 over the amount of the purchase so you can get some cash back when you buy your groceries if for some reason you can't hit the grocery store while the banks are open.
I don't even use checking withdrawal slips anymore, I just write a check for the appropriate amount.
As for convienience factor? I've found checks accepted almost everywhere I go as long as I stay in state except for restaurants, and at those places, I tend to use credit or cash anyway. (And now some restaurants take check).
The only down side is some businesses have stopped taking check and instead do a "Electronic Check Conversion" which is you authorizing the merchant to make an arbitrary sized withdrawal from your account. There is no legal fraud protection on those, so if they hand you back your check and a little credit card receipt to sign, don't do it. Announce loudly that you don't agree with such, and that you are doing a check transaction, not a electronic debit transaction. They are in theory required to notify you in advance of the transaction for any mail order transactions along with a few other regulations.
I don't even have an ATM card, never needed one. The bank should pay me for the priviledge of holding my money, not the other way around.
I've seen someone threatened with arrest for asking for the badge number of a police officer. Just because most good officers have no qualms providing badge number/ID to prove their identity doesn't mean that others won't respond with threats.
I have yet to find an area where even half of the school board members held the level of education of any of the academic subject high school teachers. As a typical example, one person I knew would regularly challenge all prospective school board members to see if they had a basic understanding of first semester calculus (which was taught to almost certainly more than 200 students per year in the school district). It was rare to find more than one canidate with that level of understanding, despite that being considered the first non-remedial math course at many colleges. It is important to realize that I am not talking about rural areas with very few members of the community who have college degrees, never mind advanced degrees. These were decent sized communities with a good sized technical workforce.
You ask who would serve on a school board if they didn't truly value education? Did you see the incidents in Kansas and the numerous other incidents at local school boards where people who clearly value a lack of education go for the school board to ensure that the children are not educated? I've seen people running for school board (and elected) who openly opposed public education as a concept, wanting to disband it. I don't think I was watching just a strange case.
Which brings up the real reason this list is worthless. We as Joe Q Public have little to no way of knowing what kind of supercomputers were delivered in secret or confidentially.
It's a brag fest for those who buy supercomputers and don't care who knows their real compute power. I have a gut feeling that is actually a rather small set of people. If you read the About, they are based on surveys of manufacturers and users.
Another note of interest is that the benchmark chosen is to solve a dense system of linear equations. Depending on what kind of problem you are buying supercomputer time for, that may not be the best benchmark. What if you are trying to solve problems based on very large sparse matrices and very fast random memory access is the defining bottleneck instead? Or another problem may require one to use integer math instead of floating point math? Such things can radically change the performance of a system. (Or to be really evil, what if your problem does not parallelize well at all? Suddenly a cluster may not be anywhere near as useful as it once was.)
As someone who has introduced a kid to both Windows and Linux on a dual boot system (he is now 5), the underlying OS didn't matter one bit. He only cared "can I play my favorite programs?". Wasn't overly difficult to convince him to slowly migrate over to gcompris. It certainly helped that gcompris has been actively developed recently.
For a young child (under age 5), a text processor like emacs or kwrite, some educational software (gcompris), and a painting program (there's one in gcompris, or use the one that comes with KDE, I have both KDE and GNOME on the system) seems to help keep my kid interested plenty. For games, he's so addicted to tux racer that I doubt I could move him off that game if I tried. (Yes, this is the commercial version.)
Yes, kids like to hit random buttons. Way back when, my kid would teach me some interesting built in macros of OS/2. Now, I set him up on a carefully constructed KDE desktop with very few icons on it (double clicking used to be a problem for him), and he is a lot less dangerous.
Another method that works well is printing it out and posting it on your cube wall so people walking by your cube can see it.
Another method that can work is by becoming a certified SAGE SA (see http://www.sagecert.org/) . One of the requirements of the cSAGE certification is to have read the code of ethics. Note, that does not mean one has to follow it, but one does need to have read it. The more cSAGE certified SAs on your team, the more your management will be aware of the certification, and the more reasonable the idea of a code of ethics will become.
As a senior SA in the team I work for, I made it clear to all the other people in my team that I expect them to be aware of the SAGE code of ethics. Again, I am not saying that they have to follow it, because I am not their supervisor. But it raises the awareness of the code of ethics.
As for making people more aware of SAGE itself, one excellent method is by showing up at SAGE local groups that are in your area. Make sure your manager knows what you are doing. Even if you are on call, frequently a deal can be worked out that someone else will cover for you while you are attending that meeting. Make sure to let your manager know that it is more than a social club, discuss some of the ideas that come up in the meetings. That way, your manager gets some constant reminders about SAGE itself, and some of the advantages that exist for it.
People have mentioned ACM codes of ethics, and ICCM codes of ethics, but there is an actual code of ethics designed for and by sysadmins: http://www.usenix.org/sage/publications/code_of_et hics.html
If you want good SAs, go look at SAGE. Or advertise for a job offering, insisting that your SAs be cSAGE certified. Don't whine about SAs, hire a good SA and make them the team lead. A good team lead SA has the authority to make things done correctly. You get the advantage of at least one good SA to do a brain transfer, the possibility that the other SAs will improve, and the advantage that the other SAs have a hard time telling the lead SA off if they want a good reference at their next job.
The other advantage of a good senior SA is to teach religion. The religion that menus belong in restaurants, and the function of a mouse is to do the nifty xterm select/paste with the middle mouse button. Contractor SAs tend to dislike automation (horrible, but true), because automation means less money in their pocket.
As for how to deal with such an environment, it does take some setup work, and a different way of thinking. I use cfengine heavily to make my life livable.
Careful work with vendors to ensure that fingerpointing won't occur is a must. Some vendors are obviously better at that than others. Usually I insist that regardless of who is at fault, every vendor involved will make appropriate experts available from the time the problem is detected until the time the problem is certified to be resolved (even if that means that the support staff hands off three or four times because their shift has ended).
Don't be surprised that SAs get annoyed with applications messing up servers. I've seen more than one case where an application was so badly coded that it brought the server to its knees. A few minor changes to the application to make it less of a brain dead design, it was humming along quite nicely without any significant impact to the server. I know I'm not the only SA to have seen things like that. I'm not saying that such is always the case, but it does happen. Maybe, just maybe, an SA knows what they are talking about when they say the application is the cause of the problem.
Where I work, I've seen the entire gamut. I've seen SAs who were great at scripting, but didn't understand simple computer security. I've seen SAs who while they weren't great scripters, I could hand them almost any troublesome problem and they would keep at it until they came up with a real solution, even if it took them a while. SAs aren't so much declining in quality as I see more and more need for SAs, and the burn out rate has gone up. Fewer and fewer old dinosaur SAs are left as they burn out and leave, and more junior SAs are too likely to burn out and leave SA before they ever become a senior SA. (It wasn't that long ago that it was common for servers to be able to be down every night for a few hours, now, it's rare that I get a server that is allowed a few hours a week of downtime.)
Back in 1991, I saw a high school student attempting to replicate a published result that looked very similar to the description in the article. The device was different, but the concept of very small weight loss of a spinning object was the same. In this variant, a high speed gyroscope was used. Turned out, the key was the electricity going through the wires to power the gyroscope and make it spin. The high schooler showed that if the electric current (going through wires) came from different perspectives, the "weight loss" could appear or disappear.
While I haven't read the extensive details, I do have to wonder if we might be seeing a similar concept here. What made me especially suspicious is the line that the effect only appears at several thousand RPMs. Turning up those RPMs means more current to the system, and more potential for side effects. Since the primary effect being looked for is very subtle, it could become significant.
One place I see a need for this kind of behavior is the user who knows enough to be interested in running linux, but isn't comfortable with programming and thus is not comfortable downloading and compiling a new kernel. The distro kernel upgrades are usually nice, but in one particular case I saw, (Mandrake 8.1), using the distro supplied tool to update the system kernel (from the distro kernel, not even a recompiled kernel from distro source) made the system unusable. I had to step in and grab a newer kernel and then compile it.
Even a reasonably seasoned user who knows what they are doing could take advantage of at least an autoconfigure type kernel utility that probed your hardware and filled in entries appropriately. For example, why should my kernel try to compile in code to fix a bug in chipset xyz if I don't have that chipset and another chipset was correctly identified in its place? It could save the trouble of looking at various outputs of lspci and similar to try to figure out exactly what hardware is on the system you are working on, something that many users may have difficulty determining.
While many of these ideas sound great in theory, the reality of administration forces many of these ideas to be abandoned.
1. Physical security. This should already be in place, and not just for credit card numbers. If you don't have adequate physical security on your servers, you have a tremendous problem. Most decent hardware has the ability to key lock access to the hard drives.
2. Network operations monitoring. Often overlooked, this feature is crucial, and contradicts some of the features others advocate of eliminating all other network connections, etc. Any service that needs to be up all the time, such as a billing server (needs to receive new records all the time for changes to billing, as well as customers querying the status of their bill, etc.), should be monitored by a dedicated operations group. The monitoring should be mostly automated (Big Brother comes to mind as a free utility, there are others both free and commercial), but with a human to evaluate the alerts you get so your admins last more than 3 months. This kind of monitoring nicely encourages a smart person to throw in some log monitoring to monitor the logs for "unusual" network activity to flag an alert. But you obviously can't yank the network cable every time you receive an extraneous port 21 connect. You need a human around to look for patterns and have the brains to know when to wake the admin and when not to. You need a dedicated staff of at _least_ 3 people (one off call, one primary on call, one secondary on call) at all times.
3. What I have found works well is to segment the process into multiple servers. Thus you can have a web server that can submit an append request to the database server, but can't do a full read, only certain types of read requests. The database server would of course be sequestered as it would hold your real data.
4. Forget locking people down to the console only. That works up until you need three people working at once to resolve an outage situation. (Don't say that doesn't happen). Instead, use an isolated network with a "hop" box that allows you to choke point access. Not the same as a firewall, but a firewall could serve the purpose as well.
5. As many others have said, make sure all traffic in and out of the box is encrypted by some method. PGP works well for some things, SSH works well for others. There are other options out there. (If you are talking to mainframes, there are some other interesting options.)
6. Good backups are essential. Sooner or later, some idiot will remove a key file at 6PM Friday, only to be noticed at 6AM Monday (or worse, not until the first of the month when you do a large monthly billing run.) There are commercial services out there that will securely store your backup tapes offsite for a given retention period. Test your backups also. On your test machine, blow away the database (prior to going live preferrably), and then restore the system from your backup tapes, and do that at least once every other year, restoring random files at least 3-4 times a year. What good is your data if it is unavailable when you need it?
7. Test machines are a must. When you want to test to see if you left open a back door, don't do it to your production system, do it to your test system that is an exact duplicate in terms of hardware type, every piece of software, and network layout (but obviously no customer data, just dummy data that emulates the structure of the database). It does no good to secure a system only to have the security compromised because you didn't anticipate a side effect that actually chokes off your primary application's access.
8. I don't know if the vendor selected uses phone validation, but if that is the route you pick, make sure your lines are configured for dial out only. Otherwise, it can leave an interesting hole that won't be revealed by even a very heavy TCP scan or lsof -i TCP
First of all, Sun CPUs are far from outstanding. Until the UltraSPARC III came out (which many people do not have yet), they were much slower than most of their workstation/server competitors. (I am assuming you are talking about workstations here because no one should be stupid enough to have a desktop environment running on a server.)
Second of all, remember that if your system is doing nothing, it is very generous to processes that would otherwise do nothing. From what you say, it sounds like some kind of network access or other runaway process is taking up your CPU time. Check your mpstat, your netstat -i, etc. to look at your bottlenecks. You could have some combination problem going on that causes a performance problem when two things are running at the same time.
As for comparison of HP's VUE (excuse me, CDE,) to GNOME, a lot of that is personal preference and habit. I used HP VUE prior to CDE coming out, and then switched to CDE and found the differences relatively minimal, even in the config files. I use GNOME at home, my wife uses KDE. My major complaint about GNOME is that I prefer the old xterm to ANY of the newer terminals I've used. (Get rid of those stupid menus, and give me back the xterm scroll bar, complete with the "strange" mouse control.) CDE's pager is mildly annoying to me, but that's because the first pager I used was similar in style to the one in FVWM and GNOME rather than CDE.
CDE is dead because of a squabble between the major CDE "partners". Ever try to get HP's CDE and Sun's version to cooperate? It's effectively impossible. The two sides are squabbling so much that CDE has not fundamentally changed since 1996 so far as I can see. It is clear that something needs to be offered soon to desktop users soon with proper commercial support. Maybe GNOME isn't the right answer yet with nautilus. (I prefer gmc myself for what little I use a file manager for.) But something does need to be done.
There is a major reason for the RC5 effort that you seem to be overlooking. Prior to the 24 hour DES crack, it was illegal to use DES in France. Shortly afterwards, France decided that DES was legal. I don't think that was coincidental. The RC5-64 crack effort is partly an effort to encourage governments to permit broader use of strong cryptography, or in the case of the United States, export stronger cryptography.
Cracking challenges also encourage people to look heavily at implementations of cryptographic algorithms to see if they can find shortcuts. That is research. It may be unlikely to result in much return, but there has been more than one example of a cryptographic implementation that had problems.