It runs 5-8% using Outlook XP accessing 14 accounts across my four servers including one account each for GMail, Hotmail, Yahoo! Mail, and CompuServe (Classic). Actually, the real CPU (whore) hog is Firefox which peaks at 58%. Even the latest and greatest (beta) version of VMWare doesn't pull that much typically unless I'm doing something rather radical at the time. You have to be careful with these numbers though as the whole system is a totally custom design with the Windows OS itself highly modified for performance as well as the hardware itself. I even went so far as to reprogram some of the values in the Northbridge and Southbridge. Even Intel wouldn't recognize what this motherboard is doing although if you study the engineering manuals the potential was always there. Then again, I have been reading engineering manuals as light reading material since I was a teenager. I'm not bragging, it's just what I consider good fun.
It's amazing the gains you can make with slight adjustments such as synchronizing the FSB to some even multiple of the the CPU clock speed. Then again, for anyone really familiar with electronic engineering, it shouldn't be. That doesn't even address some of the id10t default settings inside Windows although the amount of experimentation, man-months to say the least when you include repeated benchmarks and real-world application testing, required to find the optimal values isn't something that anyone but a prime-level geek would commit. And once you do find them you get to toss it all out with the next version but that applies to hardware as well. I'm currently in the throws of designing my next machine and finding the information to make the best design decisions ain't fun at all. Then again, I design for years (at least a decade actually) of useful life while the whole industry is geared to obsolescence as soon as it reaches the end of the production line. They can keep their upgrade treadmill, thank you. I do create and build a new machine every two years but that doesn't mean that my older machines are exactly languishing here or elsewhere. The advent of *nix in a more widespread form is very useful in that regard although optimization is a bit more difficult in many cases.
I agree and you don't have to be doing anything related to.NET to use it and get standard code as it will do compliant code just fine. I used to work professionally, especially on database-driven web designs, and have pretty much every tool imaginable and a more than a few out of any reasonable price range here and frankly ASP.NET WebMatrix is a sweet little tool. The documentation, the fact that you can get a free Wrox Press book on it, and the sheer amount of code on the ASP.NET web site are all pluses in my book. [CodeProject is another good source, btw.] Sure, it's against the orthodoxy here to say anything good about a Microsoft tool but sometimes they get it right. Okay, rarely, but it does occur. I certainly won't say that about what Office/FrontPage generate. Blech!
Then your head IT guy has selected the wrong products, probably without testing them, for your company. My system (Windows Server 2003 Enterprise as my development desktop, 2.8 GHz P4, 1 GB RAM, 256 MB nVidia FX5700,...) is wired up like a pinball machine here with everything from virus checking, spyware protection, even packet monitoring and logging and CPU load runs 1-3%, period. Heck, SQL Server doing absolutely nothing at the moment usually grabs more CPU than the rest of the system combined. What's interesting is that for individual use every product I use is free and pretty durn cheap for a business license.
The problem here, as I see it, is far too many IT types don't bother to test everything, especially vendor claims, against realistic setups before committing to the dotted line. Furthermore they don't make the case to management about the total costs, including all factors, to upper management. Lastly, upper management doesn't trust the average IT department to speak the truth, let alone deliver on their promises. I see all of this over and over in the field, when I'm called in to consult, and in the industry journals day in and day out. Actually it's sad that I have to be called in to (sometimes) give the exact same information and recommendations as an IT department simply due to the fact that I refuse to lie and I have always admitted immediately when I don't know something (but I'll go find out). I guess that makes me weird but engineering doesn't usually let you get away with BS. As I keep saying, nuclear meltdowns are sooo messy.
Frankly I don't care whether they are valid or not from a consumer standpoint. I don't deal with consumer operating systems here aside from testing against them, all my work is with enterprise level operating systems and unlike many people I do have numerous signed contracts obligating me to abide by the terms of the EULA and that doesn't even count the various NDA's I've signed. I'm also subject to audit at any time by the terms of that signed contract.
Frankly you can take whatever risks you want and it is no skin off of my nose. I prefer not to watch my computers walk out the door along with the software as evidence in a civil (or even criminal in some cases) action. I don't think any business that has also signed such contracts is going to take that risk either while they await for the courts to sort out what is legal and what is not. We have work to do. For a consumer it is an annoyance. For a business it is a work stoppage issue that can result in bankruptcy. When the Supreme Court, or a federal appeals court, shoots the whole thing down then I'll feel confident about engaging in such risks. Until then, frankly, I prefer not to deal with the issue.
I have to agree with GP. The whole long drawn out article was mostly about the install and could have been tossed off in a paragraph if I were writing the article. Hell, a sentence would do it. I have a real problem with articles that waste my time wading through fluff to get to the meat which was sadly lacking in this case. There's a lot more to SuSE than YaST, a tool I happen to really like. I'm more concerned with integration of the various packages, ease of maintenance, and issues like does it pass the "Grandmother test."
Sun is trying the same thing as well and so far doing miserably at getting people to trust them with their data and computing power. One problem is the price-point. The other has to do with systems security which, despite assertions by Sun to the contrary good but not quite good enough, IMNSHO. Where I think Google gets it right is that they are leveraging something they are durn good at, storage (these people buy hard drives in railroad car lots!;-) and retrieval and since they have that storage laying around anyway, they'll derive an additional revenue stream from the ads even if they don't charge a subscription fee. If anything, this enhances their business model rather than representing a divergent product line.
I think you are on to something here although I'd take a step farther out. One of the stumbling blocks on the web is collaboration. Sure, you can use Wiki's and a other pieces of software but that only gets you so far. If Googlebase acts as the collaborative database back-end and provides easy to use tools for the front-end, this could give Microsoft a bit of a run for the money in that market despite having Ray Ozzie as their new CTO. I think this would be the real Web 2.0, not some of the other stuff I've seen touted that way.
It would ease the creation of collaborative and derivative works and so long as those works were also stored on Googlebase, you could see an exponential growth in useful works along with a concommitment exponential growth in crap but that's the 'net for you. Still, as a multi-disciplinary synthesist something like this would be an ideal tool that, provided the meta-data were accurate, increase my capabilities here. Sadly, although I'm absolutely killer at flying the various search engines, so much of what I would like to work with is either poorly indexed (you'll find it buried on page 205 of the results despite narrow keywords) or behind closed doors which journals are notorious for and that's where so much of the cutting edge stuff is found. Lastly, Googlebase could also give an arena for self-published works, works that are not popular with either the journals or the colleagues that do the peer review. I've seen more than one journal article shot down due to heterodoxy, especially in the social sciences and medicine (both fields I've worked in).
I see it as a 'net plus (pun intended). And not just from the commercial aspects. If anything, it's an extension of GooglePrint as you've stated.
Good analysis but I'd take it a bit farther. IBM is deriving an increasing amount of revenues from its service arm. If they can push for standards across the entire storage market this will be a huge boost for that arm of the company. Many companies will turn to IBM Services for their integration and support services rather than dealing with one or more vendor contracts. True, it will almost certainly cost them a bit of their hardware market sales but more and more IBM is winding that arm of the company down except for their extreme high margin products. All in all, a huge net plus for their bottom line.
It's going pretty well here. Mostly I spend my time these days playing with betas for the big guns with the odd stint of security consulting on the side for a few select people. I also do the odd bit of consulting, mostly database engineering/re-engineering work, for various projects around the world courtesy of the 'net. It's keeping me mostly amused although I did have to heavily modify one of my systems to handle multiple boot drives to a fare-thee-well. That doesn't even count the virtual machines laying around all over the place. Oddest damn configuration you can imagine but it works.
The only other thing I have on the front burner is my attempt at creating a conceptual equivalent (threading, libraries, moderation), but much improved over Prospero of course, variant of the old CompuServe but web based and distributed. The guts are quire different, of course, as are some of the implementation aspects as I really don't feel like getting sued this week, not that I have any assets to grab. It's very SOA and by that I use the term strictly and it implements an offline reader. Right now it's.NET 2.0 based mostly due to the multiple datareader capability that it has which simplifies things immensely in the performance arena. That and the much improved XML handling features of SQL Server 2005 which I've been playing with for quite a while now (one of my beta projects). The whole implementation is pretty much in the database itself and will be distributed for the free SQL Server 2005 Express for local implementations with the presentation, outer security and validation layers in ASP.NET. You know my thoughts on security and validation so it has an inner (database) layer as well for double insurance which is why I needed triggers and stored procedures. Security has to be designed in before one line of code is committed to paper or a computer.
Which brings us back to the topic! Now that MySQL 5.0 supports triggers and the other features I need, I'll be taking a look at it again for an additional implementation. That still doesn't address the presentation layer, which would have to be something widely supported by hosting providers which probably means PHP. I'm not exactly thrilled by that prospect as PHP is a walking, breathing security hole looking for a place to happen (gets the most security notices, bar none, of anything I've ever seen and I see them all). Safeguarding the application side won't be the least bit of fun. Mono would be a nice approach but the number of hosting providers that have it can sadly be counted on the fingers of one hand. There's also the problem of when hosting providers will make the switch to 5.0 but that's a problem with.NET 2.0/SQL Server 2005 as well. Local implementations are less problematic in that regard.
I have another one in the design stage but I don't talk about it much since it's something that hasn't been done before although I do see some people (Ray Ozzie) have glimmerings in that direction but the all too standard approach they are taking is desktop/server based. I'm pretty much designing these days to the 'net. The tools which are readily, and often cheap or freely, available these days is simply staggering. And the quality improvements are proceeding at a far more rapid clip than in the proprietary arena, which is why I like to keep my hand in the beta process You can really make a difference with F/OSS and see the results. Only VMWare among all the vendors I work with seems as nimble.
As I said, I'm amusing myself which is a good thing as most people forced into retirement usually don't last more than a few (2-3) years. They declared me terminal back in '99 with 3-5 years to live. I've managed to beat that so far and I have the feeling that so long as I keep having fun (remain involved in something, I can keep trucking for a while yet.
I hate to say this to anyone, but you obviously have no clue about the architecture of Vista, Monad,.NET 2.0, nor any other aspect of Windows as it relates to the next version. Period. Case in point, how the hell is an ActiveX malware supposed to invoke MSH? First off, you'd require the id10t (l)user having to tell the system several times to run that (usually unsigned) piece of crapware. Next, it would have to literally break out of the sandbox that IE will be living in on Vista. Then it has to circumvent or get the stupid (l)user to type in their administrator password so that it can break out of the LUA lock down. Quite a feat. Sure, some (l)user will probably do it and create another zombie which will last right up until the next time the OS automagically checks WindowsUpdate, unless you really get creative on creating this piece of crapware. Fortunately most of the stuff I've seen is about a creative as I am at painting (i.e. zero).
In other words, there are going to be a lot of hurdles out there that you have to get over/around/under before you get the keys to the kingdom any more, even if the id10t (l)user runs as Administrator all the time.
This isn't to say that all is sweetness and light, but right now I'll take any bone in the security arena I can get these days as locking down and 'fixing' Windows systems ain't fun any more. Neither are Linux systems but I'll save that for another rant.
If you are operating any OS without a full system backup, your whole system is essentially a single point of failure. Furthermore, there are numerous files in the various *nixen that also act as SPOF as well and this is true of every OS that I've come across in production use. However, in Windows XP/2003 even if you don't create any backups at all, you will have at least one known good configuration which includes the registry to fall back to when something terrible happens. It was created as part of the installation process. I don't know of any 'normal' users that know how to create one but the capability is in the OS to do so when you wish and on my Windows boxes I wish every Saturday as part of my normal maintenance day for all my machines. I also make a manual backup of the registry (export to file on my file server) and tuck that in with all the other unique system configuration, database backup/log files, etc.
Your point would have been valid for Win'2K or earlier and then only for those that didn't know how to export the registry, and import it again. However, most technically inclined people knew how to do that anyway as part of the process of defragmenting the registry, a rather notorious problem at one time.
Besides, as a sysadmin, having to hunt down several hundred to thousands of config files ain't exactly fun as Linux still hasn't standardized on exactly where all those files are going to end up. Mostly yes, but there are too many oddballs out there. I know they are working on it and hopefully they'll get this resolved soon. In the meantime, a full system backup with relatively few incremental in-between is about the only way to go for now. Gee, sounds like Windows.
They all suck if you are a sysadmin. Life sucks if you're the sysadmin!
Actually CVS does exist for databases however the cost is more than a little prohibitive for your average Joe wannabe-DBA/programmer. Typically the tools (e.g. Embarcadero) run into the several thousand dollar per seat range. Their RapidSQL product, for instance, works with many existing CVS tools out there and E/R Studio, an absolutely fantastic db design tool, uses a repository with full check-in/check-out. Well worth the cost if you are serious about your tools. Heck, compared to what VS.NET 2005 will cost, full-up, the suite is on par.
This is especially interesting in light of an announcement that I came across on PhysOrg.com:
NEC DEVELOPS SPEECH-TO-SPEECH TRANSLATION SOFTWARE FOR MOBILE PHONES, October 24
NEC Corporation today announced that it has succeeded in the development of Japanese-English/English-Japanese, automatic speech translation software for single-chip multi-core processors for small devices such as mobile phones, capable of operation at high speeds with low power consumption. Full story at http://www.physorg.com/news7513.html.
Curious coincidence, neh? An application that needs a low power, multicore design and here somebody steps up to the plate and says "sure, we can do that."
It must have been really small since I didn't come across it and I read all EULA's and NDA's so I know what I'm getting myself into (legacy of my days in government service;-). Those were the days .
That is a good question. My P4 spends almost a ton of time in the field as it is also my diagnostic platform for network problems and on the spot development work. It is also highly portable as I designed it that way. So, does it qualify under the EULA? I would hazard a guess that MS would disagree rather strongly. Not that it matters much, I have licenses to burn for Office XP, not that I use it very much except to read MS documents from the MS sites.
Well if they aren't, then the unions wasted $33 million dollars during the last presidential election cycle, not counting 'volunteer' activities and in-kind contributions, for nothing. Frankly, I could care less as I consider the current crop of unions worse than useless smf would never join one. Which is why I'm not teaching in the public schools, among other reasons but that's the main one. Bunch of blood-sucking leeches.
I try not to think about Kelo as much as possible. That is most definitely not what the founding fathers intended and I can't understand where they dug that one up. If anything, Madison and Jefferson both warned most strongly about such practices.
Sorry, but I believe you to be sadly mistaken. You look at assets prior to assuming office and track them year by year on what they report (and they don't have to report everything) and you'll find radical increases in wealth. Congress-critters get all kinds of sweetheart deals including, but not exhaustive, cut-rate IPO's, subsidized/ultra-low-interest mortgages, hiring of their family members by PAC's and even their own election/re-election committees, free trips, memberships, etc. ad nauseum. Legally none of this falls into the category of bribery, ethically and morally it surely does. There is a rare member of Congress that hasn't taken advantage of at least one, if not all, of the loopholes and exceptions out there. The information is out there if you are willing to dig a bit, and you have to dig as it rarely makes the front-/main-page of the main-stream media pages. One nice thing about being (semi-)retired is I have nothing but time on my hands and I read everything that isn't nailed shut, voraciously.
Where I do agree is that the failure point is with the electorate. I'm an informed voter and I track votes as well as read the actual legalese of any constitutional amendments (should be mandatory here in California). This makes me a very rare bird indeed. Most people just go by the commercials which are almost always misleading to put it mildly. That doesn't work in a republican form of democracy. How can you know if your representative is representing you if you don't know what they've done?
Frankly, I don't know how you would fix it with the current crop of voters. I do know that for future voters you would have to literally destroy every school of education in the country, fire all of the teachers, and then start over in your hiring as the school system itself is completely incapable of turning out an educated populace, especially now that we've switched from an industrial age economy to an information/services economy. Without a broad liberal arts background starting in elementary school, you are completely unequipped to understand the context of the issues let alone place them in a historical or philosophical perspective. And that doesn't even address the requirements for understanding scientific issues.
As you can tell that's a hot button issue with me. I've been teaching for the last 33 years at technical schools and the university, on a formal and an informal basis, and frankly in any other profession what they do in our schools would be liable to a malpractice suit if they were practicing medicine.
Frankly I'm extremely pessimistic that the voters will ever smarten up. de Tocqueville had it right.
I suppose it's pointless to bring some facts into this discussion but due to a 1 GB download, I have time to waste anyway. The reason why the Navy is bewildered is that: (1) no one has seen any evidence whatsoever that whales are affected by mid-range active sonar before and we've been using it for over fifty years; (2) mid-range sonar is not used in anything other than deep-water, due to reverberation effects, far off coast-lines so how it could affect breeding areas or beachings is beyond reason; (3) sonarmen (and now women) happen to like listening to whales as much as the next marine biologist and standard practice has always been to let the biologicals (as we call them) clear an exercise area before starting an exercise. Heck, I loved going up to combat and listen to the whales and other biologicals myself when I had free time, which wasn't very often.
For those that don't follow the NRDC, they do this about annually, as someone pointed out, around fund-raising time, and the US Navy is a favorite for their cross-hairs. Personally, I have more respect for Greenpeace (and was a member at one time) than the NRDC. At least Greenpeace is run by real activists with occasional flashes of real science behind their positions rather than a collection of lobbyists/lawyers with a collection plate in front of them. Not that Greenpeace wins any prizes but I won't go there today. The real people to talk to about this issue all pretty much work over at the Wood Hole's Institute and I've never seen them say anything about this subject. Those people live and breath marine biology.
One other point. One of the things that the US Navy likes to do, in addition to our regular duties, is to track the migratory patterns of the various species of whales. Actually, we collect a lot of data while out there, not just on whales; it gives us something to do on our watches. Now blasting whales with sonar wouldn't be exactly bright if you are collection scientific data on them at the same time, would it?
EVERYTHING Congress (and the executive branch, as well) does is done as favors to big business.
I must respectively disagree as your reply isn't inclusive enough. I would rather say that everything Congress and the Executive branch does is in favor of whomever gives them the most benefits be it in the form of donations for their campaigns, free/subsidized trips or other goods and services, and similar items. The source can be big business, labor unions (e.g. AFL/CIO, AFSCME), professional organizations (e.g. trial lawyers association), and even individuals (e.g. George Soros). We have the best politicians money can buy.
On the plus side I will say that the level of transparency of corruption is much higher than I've encountered elsewhere in the world and you, usually, don't get killed for investigating who bought which politician, which I've seen before.
Sheesh! Hell, I was doing business process automation back in the early '80's so I sure as frag got prior art on that one (and official government documentation to back it up). I don't know about the AI aspect although many elements of my applications incorporated self-adjusting modeling, learning, and automatic prediction (even what would be later called neural nets). I'd like to see those patents.
Slashdot Mirror
It's amazing the gains you can make with slight adjustments such as synchronizing the FSB to some even multiple of the the CPU clock speed. Then again, for anyone really familiar with electronic engineering, it shouldn't be. That doesn't even address some of the id10t default settings inside Windows although the amount of experimentation, man-months to say the least when you include repeated benchmarks and real-world application testing, required to find the optimal values isn't something that anyone but a prime-level geek would commit. And once you do find them you get to toss it all out with the next version but that applies to hardware as well. I'm currently in the throws of designing my next machine and finding the information to make the best design decisions ain't fun at all. Then again, I design for years (at least a decade actually) of useful life while the whole industry is geared to obsolescence as soon as it reaches the end of the production line. They can keep their upgrade treadmill, thank you. I do create and build a new machine every two years but that doesn't mean that my older machines are exactly languishing here or elsewhere. The advent of *nix in a more widespread form is very useful in that regard although optimization is a bit more difficult in many cases.
Actually given the fact that Screem is text-editor based I'd call it a competitor to HomeSite+ not a competitor to DW.
I agree and you don't have to be doing anything related to .NET to use it and get standard code as it will do compliant code just fine. I used to work professionally, especially on database-driven web designs, and have pretty much every tool imaginable and a more than a few out of any reasonable price range here and frankly ASP.NET WebMatrix is a sweet little tool. The documentation, the fact that you can get a free Wrox Press book on it, and the sheer amount of code on the ASP.NET web site are all pluses in my book. [CodeProject is another good source, btw.] Sure, it's against the orthodoxy here to say anything good about a Microsoft tool but sometimes they get it right. Okay, rarely, but it does occur. I certainly won't say that about what Office/FrontPage generate. Blech!
The problem here, as I see it, is far too many IT types don't bother to test everything, especially vendor claims, against realistic setups before committing to the dotted line. Furthermore they don't make the case to management about the total costs, including all factors, to upper management. Lastly, upper management doesn't trust the average IT department to speak the truth, let alone deliver on their promises. I see all of this over and over in the field, when I'm called in to consult, and in the industry journals day in and day out. Actually it's sad that I have to be called in to (sometimes) give the exact same information and recommendations as an IT department simply due to the fact that I refuse to lie and I have always admitted immediately when I don't know something (but I'll go find out). I guess that makes me weird but engineering doesn't usually let you get away with BS. As I keep saying, nuclear meltdowns are sooo messy.
Frankly you can take whatever risks you want and it is no skin off of my nose. I prefer not to watch my computers walk out the door along with the software as evidence in a civil (or even criminal in some cases) action. I don't think any business that has also signed such contracts is going to take that risk either while they await for the courts to sort out what is legal and what is not. We have work to do. For a consumer it is an annoyance. For a business it is a work stoppage issue that can result in bankruptcy. When the Supreme Court, or a federal appeals court, shoots the whole thing down then I'll feel confident about engaging in such risks. Until then, frankly, I prefer not to deal with the issue.
I have to agree with GP. The whole long drawn out article was mostly about the install and could have been tossed off in a paragraph if I were writing the article. Hell, a sentence would do it. I have a real problem with articles that waste my time wading through fluff to get to the meat which was sadly lacking in this case. There's a lot more to SuSE than YaST, a tool I happen to really like. I'm more concerned with integration of the various packages, ease of maintenance, and issues like does it pass the "Grandmother test."
Sun is trying the same thing as well and so far doing miserably at getting people to trust them with their data and computing power. One problem is the price-point. The other has to do with systems security which, despite assertions by Sun to the contrary good but not quite good enough, IMNSHO. Where I think Google gets it right is that they are leveraging something they are durn good at, storage (these people buy hard drives in railroad car lots! ;-) and retrieval and since they have that storage laying around anyway, they'll derive an additional revenue stream from the ads even if they don't charge a subscription fee. If anything, this enhances their business model rather than representing a divergent product line.
It would ease the creation of collaborative and derivative works and so long as those works were also stored on Googlebase, you could see an exponential growth in useful works along with a concommitment exponential growth in crap but that's the 'net for you. Still, as a multi-disciplinary synthesist something like this would be an ideal tool that, provided the meta-data were accurate, increase my capabilities here. Sadly, although I'm absolutely killer at flying the various search engines, so much of what I would like to work with is either poorly indexed (you'll find it buried on page 205 of the results despite narrow keywords) or behind closed doors which journals are notorious for and that's where so much of the cutting edge stuff is found. Lastly, Googlebase could also give an arena for self-published works, works that are not popular with either the journals or the colleagues that do the peer review. I've seen more than one journal article shot down due to heterodoxy, especially in the social sciences and medicine (both fields I've worked in).
I see it as a 'net plus (pun intended). And not just from the commercial aspects. If anything, it's an extension of GooglePrint as you've stated.
Good analysis but I'd take it a bit farther. IBM is deriving an increasing amount of revenues from its service arm. If they can push for standards across the entire storage market this will be a huge boost for that arm of the company. Many companies will turn to IBM Services for their integration and support services rather than dealing with one or more vendor contracts. True, it will almost certainly cost them a bit of their hardware market sales but more and more IBM is winding that arm of the company down except for their extreme high margin products. All in all, a huge net plus for their bottom line.
The only other thing I have on the front burner is my attempt at creating a conceptual equivalent (threading, libraries, moderation), but much improved over Prospero of course, variant of the old CompuServe but web based and distributed. The guts are quire different, of course, as are some of the implementation aspects as I really don't feel like getting sued this week, not that I have any assets to grab. It's very SOA and by that I use the term strictly and it implements an offline reader. Right now it's .NET 2.0 based mostly due to the multiple datareader capability that it has which simplifies things immensely in the performance arena. That and the much improved XML handling features of SQL Server 2005 which I've been playing with for quite a while now (one of my beta projects). The whole implementation is pretty much in the database itself and will be distributed for the free SQL Server 2005 Express for local implementations with the presentation, outer security and validation layers in ASP.NET. You know my thoughts on security and validation so it has an inner (database) layer as well for double insurance which is why I needed triggers and stored procedures. Security has to be designed in before one line of code is committed to paper or a computer.
Which brings us back to the topic! Now that MySQL 5.0 supports triggers and the other features I need, I'll be taking a look at it again for an additional implementation. That still doesn't address the presentation layer, which would have to be something widely supported by hosting providers which probably means PHP. I'm not exactly thrilled by that prospect as PHP is a walking, breathing security hole looking for a place to happen (gets the most security notices, bar none, of anything I've ever seen and I see them all). Safeguarding the application side won't be the least bit of fun. Mono would be a nice approach but the number of hosting providers that have it can sadly be counted on the fingers of one hand. There's also the problem of when hosting providers will make the switch to 5.0 but that's a problem with .NET 2.0/SQL Server 2005 as well. Local implementations are less problematic in that regard.
I have another one in the design stage but I don't talk about it much since it's something that hasn't been done before although I do see some people (Ray Ozzie) have glimmerings in that direction but the all too standard approach they are taking is desktop/server based. I'm pretty much designing these days to the 'net. The tools which are readily, and often cheap or freely, available these days is simply staggering. And the quality improvements are proceeding at a far more rapid clip than in the proprietary arena, which is why I like to keep my hand in the beta process You can really make a difference with F/OSS and see the results. Only VMWare among all the vendors I work with seems as nimble.
As I said, I'm amusing myself which is a good thing as most people forced into retirement usually don't last more than a few (2-3) years. They declared me terminal back in '99 with 3-5 years to live. I've managed to beat that so far and I have the feeling that so long as I keep having fun (remain involved in something, I can keep trucking for a while yet.
In other words, there are going to be a lot of hurdles out there that you have to get over/around/under before you get the keys to the kingdom any more, even if the id10t (l)user runs as Administrator all the time.
This isn't to say that all is sweetness and light, but right now I'll take any bone in the security arena I can get these days as locking down and 'fixing' Windows systems ain't fun any more. Neither are Linux systems but I'll save that for another rant.
Your point would have been valid for Win'2K or earlier and then only for those that didn't know how to export the registry, and import it again. However, most technically inclined people knew how to do that anyway as part of the process of defragmenting the registry, a rather notorious problem at one time.
Besides, as a sysadmin, having to hunt down several hundred to thousands of config files ain't exactly fun as Linux still hasn't standardized on exactly where all those files are going to end up. Mostly yes, but there are too many oddballs out there. I know they are working on it and hopefully they'll get this resolved soon. In the meantime, a full system backup with relatively few incremental in-between is about the only way to go for now. Gee, sounds like Windows.
They all suck if you are a sysadmin. Life sucks if you're the sysadmin!
Completely off-topic, but you wouldn't have been on CompuServe for years and years, would you? -Brian J. Bartlett
Actually CVS does exist for databases however the cost is more than a little prohibitive for your average Joe wannabe-DBA/programmer. Typically the tools (e.g. Embarcadero) run into the several thousand dollar per seat range. Their RapidSQL product, for instance, works with many existing CVS tools out there and E/R Studio, an absolutely fantastic db design tool, uses a repository with full check-in/check-out. Well worth the cost if you are serious about your tools. Heck, compared to what VS.NET 2005 will cost, full-up, the suite is on par.
Curious coincidence, neh? An application that needs a low power, multicore design and here somebody steps up to the plate and says "sure, we can do that."
It must have been really small since I didn't come across it and I read all EULA's and NDA's so I know what I'm getting myself into (legacy of my days in government service ;-). Those were the days .
That is a good question. My P4 spends almost a ton of time in the field as it is also my diagnostic platform for network problems and on the spot development work. It is also highly portable as I designed it that way. So, does it qualify under the EULA? I would hazard a guess that MS would disagree rather strongly. Not that it matters much, I have licenses to burn for Office XP, not that I use it very much except to read MS documents from the MS sites.
Well if they aren't, then the unions wasted $33 million dollars during the last presidential election cycle, not counting 'volunteer' activities and in-kind contributions, for nothing. Frankly, I could care less as I consider the current crop of unions worse than useless smf would never join one. Which is why I'm not teaching in the public schools, among other reasons but that's the main one. Bunch of blood-sucking leeches.
I try not to think about Kelo as much as possible. That is most definitely not what the founding fathers intended and I can't understand where they dug that one up. If anything, Madison and Jefferson both warned most strongly about such practices.
Where I do agree is that the failure point is with the electorate. I'm an informed voter and I track votes as well as read the actual legalese of any constitutional amendments (should be mandatory here in California). This makes me a very rare bird indeed. Most people just go by the commercials which are almost always misleading to put it mildly. That doesn't work in a republican form of democracy. How can you know if your representative is representing you if you don't know what they've done?
Frankly, I don't know how you would fix it with the current crop of voters. I do know that for future voters you would have to literally destroy every school of education in the country, fire all of the teachers, and then start over in your hiring as the school system itself is completely incapable of turning out an educated populace, especially now that we've switched from an industrial age economy to an information/services economy. Without a broad liberal arts background starting in elementary school, you are completely unequipped to understand the context of the issues let alone place them in a historical or philosophical perspective. And that doesn't even address the requirements for understanding scientific issues.
As you can tell that's a hot button issue with me. I've been teaching for the last 33 years at technical schools and the university, on a formal and an informal basis, and frankly in any other profession what they do in our schools would be liable to a malpractice suit if they were practicing medicine.
Frankly I'm extremely pessimistic that the voters will ever smarten up. de Tocqueville had it right.
Correction to my post: Woods Hole Institute. I need some sleep!
For those that don't follow the NRDC, they do this about annually, as someone pointed out, around fund-raising time, and the US Navy is a favorite for their cross-hairs. Personally, I have more respect for Greenpeace (and was a member at one time) than the NRDC. At least Greenpeace is run by real activists with occasional flashes of real science behind their positions rather than a collection of lobbyists/lawyers with a collection plate in front of them. Not that Greenpeace wins any prizes but I won't go there today. The real people to talk to about this issue all pretty much work over at the Wood Hole's Institute and I've never seen them say anything about this subject. Those people live and breath marine biology.
One other point. One of the things that the US Navy likes to do, in addition to our regular duties, is to track the migratory patterns of the various species of whales. Actually, we collect a lot of data while out there, not just on whales; it gives us something to do on our watches. Now blasting whales with sonar wouldn't be exactly bright if you are collection scientific data on them at the same time, would it?
On the plus side I will say that the level of transparency of corruption is much higher than I've encountered elsewhere in the world and you, usually, don't get killed for investigating who bought which politician, which I've seen before.
Sheesh! Hell, I was doing business process automation back in the early '80's so I sure as frag got prior art on that one (and official government documentation to back it up). I don't know about the AI aspect although many elements of my applications incorporated self-adjusting modeling, learning, and automatic prediction (even what would be later called neural nets). I'd like to see those patents.
It may be a legacy dinosaur but I don't think that fact has reached its brain yet ;-).