I agree that peer "Code Review" is a very important process. We've implemented a code review where I work, and it has been very successful. In fact, I'd recommend using a similar process to developers in almost any situation.
Attendance to our code review is not mandatory (unless you are submitting code), but we really encourage any developer who isn't in "crunch time" on their particular project to attend. Our code review really incorporates all these ideas: 1) Peer review of code going into the testing environment. 2) "Idiot check" testing by peer developers to try and break the code while still in the development environment. 3) A place to showcase new, cool code - code that accomplishes a new goal, or accomplishes an old goal in a significantly better way. 4) A place to describe new processes and/or methodologies that are being put into place for development. (For instance, one of the first things I did when I arrived was spearhead the use of source control software through this process.) 5) A place to fawn over and discuss the latest software and hardware gagets. (mimio started showing up everywhere after one of these)
The only thing that does not go into code review that makes it into the testing environment is "copy n' paste" code - code that is almost identical to another piece that has already been written. (A lot of our ASP pages are like that - call some stored procs, has a form, dropdowns)
I can honestly say that code review has made me a better programmer. First of all, it puts extra sets of eyes on your code - and that makes a difference - I guaruntee people will spot things you did wrong. Secondly, it gives me a chance to see other code that is new and well written. Then I can incorporate those ideas later on in my own code.
Furthermore, it prevents inferior code from going into the user acceptance testing environment and, consequently, the production environment. Every company has a few developers who write second rate or below standard code, and this weeds that code out before it can get in and screw up or slow down the system.
Now, obviously, every developer does not have time to step through the logic on every piece of code that comes through. Thats not really the idea though - thats the developers job. Mostly readability, the use of standard naming conventions, and obvious problems are the sort of thing that is looked for. "Can I debug and/or modify this code if I need to without your help?" is a question I ask myself when reviewing. Generally, the only stuff looked at in heavy detail are pieces that are experimental, bottlenecks, or intricate.
It only takes up about 2 hours of my week, and it makes all the code that goes into our system significantly better, as well as improving our programming skills. All in all, I'd say its a very worthwhile use of our time.
I've done a lot of research into the HTPC (Home Theater PC) market lately. I've read up on XP Media Center quite a bit. I don't think its an ideal solution, but I might buy it if they sold it as a standalone product rather than only with crappy OEM machines. Let me clear some things up for you:
Microsoft disabled the DRM "features" that were originally in the Win XP Media Center. This was done just before it was officially released, and only because of an outcry from the OEMs that were trying to sell these PCs, because those OEMs know that if the recorded videos were all DRMed up, nobody would ever have a reason to choose such a system at approx. $1200 over a $200 Tivo.
The "guide" you refer to is used by just about all the computer TV-tuner solutions out there. It is currently free. How long it will remain so, I cannot say, but it is not controlled by Microsoft; unless they built their own, which is possible - I haven't seen any of these Media Center boxes in person.
Adding HDs is no problem. Make no mistake, this is a regular PC running Windows XP - its just that Windows XP has had an extra layer slapped onto it so that its easier to access all the media functions from a single location.
The major downfall at this point of Media Center in my opinion is twofold:
1) OEM only. Until MS drops this policy, this OS will not become widespread. Most of us geeks who are really jonesing for this aren't willing to shell out $1200 for, esentially, an operating system in a fancy box. Lets see: don't like the case, don't like the power supply, don't like the motherboard, don't like the sound card, needs a bigger hard drive - you get the idea.
2) Proprietary video format. Yeah, currently the Media Center PC only records in a lame ass MS proprietary video format. (Again, haven't seen it in person, so I can't say this for 100% sure, but all I've read indicates this is still true.) I'm sure this will be circumvented, and some haxor will come out with a method of converting it to MPEG 2 or somesuch, but it still sucks.
So, basically, you'd want to buy one of these things under the following conditions: 1) You aren't a *nix zealot. (Unless you are a hypocritical *nix zealot) 2) You're in the market for a new PC. 3) You don't care about the specific parts in said PC. 4) You don't care about anyone using a non-MS operating system being able to play recorded files you send them. 5) You don't care that you will now have a computer in your living room, and all the ugliness/noisiness that entails.
The latest issue of PC World hit the nail on the head, when they said (parphrasing): "this is really only attractive to people living in either a tiny apartment or a college dorm." In other words, people who already have to deal with a computer being in their living room in the first place.
All kidding aside, its obvious to me that at least one of the following must be true:
1) This paper was written purposely to fool an ordinary reader (such as a journalist) into believing that human influence extends further than it really does.
2) The journalists who wrote the headlines and articles describing this paper clearly intended to fool readers into believing that the paper describes human influence accurately, even though it does not.
3) Journalists are idiots who don't know their head from their ass, and are incapable of descibing a paper such as this accurately.
I say this because the headlines and articles all describe this paper with inaccuracy, and the fault must lie somewhere. Most people posting here obviously believe its option #1. Its obvious to me that you either believe that journalists are either environmentally biased or complete morons (Options #2 and #3, respectively).
Frankly, I believe all 3 options are true to some extent.
Not sure why this keeps getting modded up in every single thread..
Moderators, I think its time we declared this the new All Your Base/Hot Grits/Natalie Portman/etc "Slashdot catch phrase" (TM). It should be modded down as such. After all, its only funny the first couple hundred times.
Please check the following link for information on how Slashdot catch phrases (TM) and other Memes work: http://totl.net/Cool/
You may now continue your regularly scheduled productivity draining activity.
Take a look at this website, it'll tell you the current locations and distances of spacecraft on solar system escape trajectories. (Pioneer 10, Pioneer 11, Voyager 1, Voyager 2)
Pioneer 11 is actually the least distant of all these spacecraft - and is also essentially a manmade rock at this point - its no longer operational.
Voyager 1 owns the record for farthest out, and will for the forseeable future, as it is travelling considerably faster than the others relative to the sun.
Also interesting is that the Voyager twins will be the two farthest spacecraft out once Voyager 2 passes Pioneer 10.
This should be in late November 2021 by my admittedly crude calculations. Take Pioneer 10's 13.986 AU lead, divide it by the 0.727 AU V 2 is gaining per year on P 10, and you get approx. 19.24 years. Add that to todays date and you arrive somewhere in late Nov. 2021 - I'm not going to bother with the exact date.
I'd like to make some comments from the point of the research I've done so far on this subject, in relation to some of the things you've said.
First, on hard drive failure: Hard drives fail. This seems obvious to me, I was a network admin in college and I saw a lot of hard drives fail. Pity the poor fool who tries to transport them.. click click, click click. My Tip: Burn. CD burners are cheap, reliable, and the media is super cheap. Yeah, an entire movie won't fit on on CD. But you won't be playing from the CD, the idea is merely to have a copy on permanent media in case of HD failure - then put it back on the new HD when you get it. So, make sure you have your media box hooked to your LAN and copy those files to your CD burning machine. No burner? Buy one. - Note: My Pine drive shipped with Nero full version, thus I recommend buying one, because otherwise I would have bought it seperately anyway.
Noise: Computers are noisy. This is probably the biggest hurdle to overcome in a entertainment center pc. My Tip: abandon convention. To combat noise, I'm looking at using a specialized watercooled case. I realize this case doesn't fit the traditional entertainment device size/configuration - but I've abandoned that notion and just committed myself to buying the quietest setup I can find and figuring out a way set up my entertainment center around it so that it doesn't look out of place. I have to admit, at first, a "dumb terminal" type setup seems like a good idea, but most of the stuff has to run from the client, not the server. You basically aren't buying yourself any help because your powerful video card and processor have to be on the client anyway, and those cause heat - so even a dumb terminal is going to be noisy.
TIVO Functionality: A must-have. Clearly an entertainment PC is only going to be more than a fancy looking DVD player if you include video capture and playback. My Tip: ATI All-in-Wonder. The ATI All-in-Wonder is still the best video capture card on the market. Note that it comes with a remote. NVidia has one now, but reviews are not as glowing. Program guides for both are free.
Output: TV resolutions aren't very good. My research suggests that even new, fancypants TVs resolution isn't good enough to use for a computer. My Idea (not advice): projector. Currently I'm considering buying a high resolution computer projector for about the same cost as a very nice TV (a.k.a $2,000-$3,000). I've seen one of these at one of the places I've worked, and I was impressed. It showed both a 1024x768 computer screen and DVD playback very well. (Comon, I *HAD* to play a DVD on it!) The only issue here is obviously light - projection does not work well in a room with a lot of light. Even though otherwise this is an ideal solution (huge screen, relatively low cost, high res) - I'm not sure I can live with the tradeoff. I imagine for normal TV viewing I will still go with the regular TV I already own. My super-sneaky idea for a "close" monitor - build a flap to come up closer to the projector. I doubt the text would be readable, though. I'm still mulling over the possibilities.
Alright, maybe the guide does suck. Why not list: A) Kits you think are better or B) What parts you would use that are better or both.
Its easy to say "this sucks" and leave it at that. The rest of us would like to know, if there are better options, what are they? The whole point of the kit is to make this sort of thing accessible to geeks like me who want to do a watercooling solution, but don't have the time to learn everything there is to know about pumps, tubing, etc - and don't want to spend the additional time trying to locate the best parts for the job. What we need is a list of parts, where to get them, and if there are tricky parts to the job, a short guide. I'd settle for a list of the best parts suited for this task though.
A case with a lock on it? - You'll have to crack into it without a metal cutting saw.. and thats gonna be noisy and take some time. (Not to mention leave a mark.) Note also that most manufacturers sell cases with BIOS level intrusion detection built in if you want it.
I applaud the efforts of junior MacGyvers, but if you really want to be secure, there are obviously better solutions.
Speaking of computer security, that reminds me of the time when the CS department at the University I went to got a bunch of brand new lab machines. They all had intrusion detection, which we CS dept. admins thought was pretty cool. We told the campuswide IT guys that we needed them secured in place. They dragged their feet on it. A month later, the CS department threw a Lan party in the same building (though not in the same room) and some enterprising students used it to cover the theft of 4 of the new lab machines. Security wire and cameras were in the room within a week. As far as I know the stolen computers were never recovered. We took small solace in the knowledge that the computer beeps at them and displays a brief annoyance message now before booting the OS. That is, as long as they opened the case and didn't flash the BIOS.
In psuedo code, I might agree with you. In a specific language, requiring it compile, no. Yes, a good programmer should know basic concepts, but more importantly, a good programmer should be able to use basic concepts in a language independant manner. Why would anyone want a programmer to know how to code a linked list? Its a waste of time. No one is gonna code it from scratch, its a trivial and demeaning question. I know if I went into an interview and someone asked me for code like that, I'd tell them straight up that I'd look that up in a book or online, and that it'd be a waste of my time to write it. Maybe if you're hiring greenhorns straight out of school you might ask a dumb question like this just to make sure they aren't completely clueless, but for experienced programmers its frankly a poor test to find a good problem solver. Really, in my opinion, it shows a lazy interviewer who hasn't bothered to come up with good questions.
I'd argue that you are not only eliminating the worst candidates with that type of test, but also the best. The best programmers do not remember code in a specific language. The best programmers do not waste valuable "brain space" remembering trivial crap like linked lists that none of us would ever write in practice (since its been written a bazillion times anyway).
Writing code is not done in a sterile, non-networked environment without books. We do not code on the top of Mount Everest. Good programmers certainly will not waste a companies time and money by writing a linked list from scratch. Knowing how to write code in a specific language is one thing, knowing how to write some trite piece of code that you haven't written for 5, maybe 10 years is another.
I say, if you're going to ask for code, ask the interviewee beforehand what types of things they've been coding recently and ask a code question along those lines. Do a little work yourself instead of asking some lame ass question that no one would have any good reason to remember the answer to. Hope you like those greenhorns or anal rote memorizers you've been hiring. They must be either close enough to having been in school or a one language pony to remember useless junk like exactly how to write a linked list.
Frankly, I'd be more apt to hire a candidate who told me writing a linked list was a stupid question and that he'd just look up sample code than one who wrote it perfectly. I don't want droids working with me, I want problem solvers. That is what programming is all about, after all.
In short, unless you're hiring for the position of linked list expert extrodinare, asking for on the spot compilable code for a linked list is most likely a waste of time for the interviewer and the interviewee.
I have to wonder, was exclude all stories by author implemented solely because of Katz? If it predates Katz, certainly it was a prescient decision.
At the very least, I would be willing to bet that half of all slashdot user accounts were created to eliminate Katz stories from showing up on the main page. I know mine was. Maybe this should be the next poll?
Did you create your slashdot account so that you could eliminate Katz stories? *Yes *No *I don't have a slashdot account you insolent clod *I just use cowboyneals account
And what are the percentages on which authors are excluded by slash user accounts? I'll bet Katz is winning by a landslide. 70%? 80%? More?
On a side note, you might wonder why I'm replying to a Katz article if I exclude his stories. Different computer, wasn't logged into slash as me.
Oh, also, I wanted to add a me too section. Quite a number of posts have pointed out an earlier slash article disagrees with Katz about the magic of myth bullcrap and points out the true origins of Star Wars in sci fi pulp fiction. To those posts I say: Me too!
He's probably not a card-carrying, slashdot reading member of the lumpenintelligentsia such as yourself though, so you really don't care.
I do care, thats why I my post tries to point out that the target of this trap is often poor people who often feel they have no alternative but to steal. Seems you're the one who doesn't care:
Seems now the refrain goes: "Give me your tired, your poor..." - so I can throw them in jail.
Only if they're car thieves.
I'm not even going to comment on your irrelevant stuff about Abscam. A program busting rich congressmen for taking bribes and one busting poor people for stealing cars aren't even similar in the terms of my argument.
Take some advanced college courses in criminal justice or sociology, then you might actually learn something and not spout mindless crap like this on/.
I did learn something in those classes. This isn't mindless crap, but I'd argue your double talking "you don't care about them, throw the bastards in jail" post is.
Who cares? You SHOULD care. The police are busy wasting your tax dollars and mine in order to put more poor people in jail. Meanwhile, corprate American criminals continue to rape and pillage the country while you and I pay for it. What I'm saying is, shouldn't we be spending money to trap corporate theives costing us billions than poor people costing us millions? Shouldn't we be after people who steal out of greed rather than those who steal out of need?
I haven't watched the X-Files for at least two seasons, after the tremendous sucking of three seasons ago. The fact that anyone is still watching that show astounds me.
Pulling this sort of crap with the best characters the show had come up with for years doesn't surprise me.
The show really should have ended with the movie, it was showing signs of sucking then. Afterwards, the show took a dive into a soap opera set in a parinoid delusion of space alien invasion. Yeah, I know the show always had that bent, but it went off the deep end after the movie. We begged them, go back to the monster of the week! But alas, they didn't listen, and alienated (literally) many of their fans. All the mystery was removed from the mysterious characters, at least, those that were still alive after the movie went on an important character killing binge. With the skepticism removed from the Sculley character, the show had no way to ground itself in some semblance of reality. Really, there was no one and nothing left for the show to run on.
I find it merciful that the show only has 4 episodes to go. X-Files, you once were a great show, but you went on far beyond your prime, and you've tarnished your image in front of many of your fans.
The preceding opinions are my own and do not neccessarily reflect those of geeks whose eyes towards the X-Files glazed over long ago.
How many car alarms did you ignore today? 1? 10? More?
Lets examine the situation here: A "late model sedan" a.k.a late 90s Honda is put in a poor, run down neighborhood and left with the doors unlocked.
Poor people recognize this car: A) Does not belong there. (Despite what the article says) B) Is not owned by anyone they know. C) Is unlocked. (By trying the door)
For a person walking by this car represents easy money. If you were poor, would you walk by $10,000+ sitting on the street?
Granted, most of the people who steal this car are going to already be car theives. But you have to admit there is a distinct possibility that this car could tempt someone who was not a car thief to become one.
Seems now the refrain goes: "Give me your tired, your poor..." - so I can throw them in jail.
Take an introductory college course in Criminal Justice or Sociology. It'll really open your eyes.
I'll leave you with this: Who steals more: poor people scraping a living, or rich people in the boardrooms of corporate America? What theft has more impact on the lives of most people?
On an oddly related point, I saw a thing on TV recently about new inventions and some contest run every year to give an award to the most promising new invention. This one wasn't voted most promising, but it was cool:
One guy invented a speech feedback system that picked up your voice in a headset microphone, changed it slightly, then fed it back into earphones you had on. He previously had a serious studdering problem, but this device allowed him to speak almost perfectly after having used it repeatedly to practice speaking. He demonstrated that the device could also be used to make a person speak authoritatively, cheery, fast, slow, and various other ways as well by changing the feedback that they were hearing.
Pretty amazing, how the brain would change the voice to match what it thought was normal almost immediately. People wouldn't have any problem adjusting voice inflection to match the device assuming it provided feedback.
I'll try and give a few reasons for the eventual demise of Solaris in the following long winded spiel:
First off, I don't subscribe to the imminent demise of Unix conspiracy theory. Various flavors of Unix have a strong hold on the server market, and with good reason. I don't see this changing. What I do see changing is the market share of Suns Solaris platform.
Ever tried to install and run Solaris on PC hardware? Trust me, its painful. Now, I know what your going to say: a lot of servers don't run on PC hardware, so having a working install and running well on such hardware is irrelevant. Thats where I might begin to disagree with you.
In the next 5-10 years, I expect to see a serious movement towards _networked_ distributed computing in the server market. I know this is in use in some applications now, but I think we've only seen the tip of the iceberg. Distributed power + useful real world application (aka not Seti) = market bonanza. And with computing distributed among many machines in a network environment, I expect to see a movement gravitating towards the use of cheaper PC hardware over massive proprietary server machines.
Think about it: what sort of things are most servers doing now? Serving data, serving web pages, handling user accounts, backing up data, and maybe running some programs of some kind. Now, you're probably saying, its that last part that the big dedicated servers are all about. True, but consider this, what if each of the programs were running on its own PC server? Wouldn't that be the same as running on a small portion of a large servers CPU cycles? And if its not, then we're probably mostly out of Solaris's market anyway. Do we need that massive server if we can simply make a lot of little servers look like a big server to anyone who is using it? There is potential here to make data backups mostly irrelevant, since each PC server holds an overlapping piece of the total data, the loss of one or even a few machines doesn't matter. The emergence of IDE RAID makes this an even more distinct possiblity, because of protection at both the machine and the network levels. There is potential to scale properly and closely to a clients exact needs, causing a client to spend a lot less money on a computing platform. Finally, there is potential for maintainence costs go down, due to the lack of needing someone with special skills to handle specialized hardware. There are disadvantages, I won't deny that. However, I think when it is properly done, the advantages of this type of server environment will vastly outweight the disadvantages.
And as for that imminent demise of Unix thing, I think the future is rather bright, because I think it will be some flavor of Unix that brings this sort of technology to the masses. If I'm right, that flavor of Unix will not be Solaris.
Please, there weren't enough obscure acronyms in this piece for your average government agency. We need more. Have someone get on this ASAP. We need more acronyms PDQ. We need to assign a DOAP and make sure they PATFT. That way we'll all be MHIB. Clearly LIAP for posts or we'd have LODT. So lets KUTGW with OAP and we'll all be VAFWWH.
I consider myself a tech-head, and if I can't make sense of a tech article at a glance after getting a Bachelors in Computer Science, something is wrong. I don't even know if I'm interested in this article. It has something to do with the internet, emergencies, and 9/11; and the rest is friggen jibberish. To add insult to injury, michael the slashdot moderator adds an unrecognizable acronym of his own!
PSTN? GETS? IEPREP? Not to mention the slightly better known RFC and IETF? This is crazy. IMHO, I shouldn't have to follow a link just to find out WTF the article is about. These kind of posters need to STFU or slashdot will be a FUBAR POS that just wastes my time.
DOAP: Designated Obscure Acronym Poster PATFT: Post All The Friggen Time MHIB: Much Happier I Bet LIAP: Length Is A Priority LODT: Lots Of Descriptive Terms KUTGW: Keep Up The Good Work OAP: Obscure Acronym Posts VAFWWH: Very Appreciative For What We Had
That was the idea, when Java first arrived on the scene..
Since Java was platform independent, you could write a Java application, and it would run the same no matter what was running underneath, be it Windows, Linux, or what have you. This meant that if a lot of truly useful software had been written in Java, Linux would start to become a seriously viable alternative, because it could run the the truly useful software just as easily as windows. As the first poster in this thread astutely observed, MS and Sun combined their powers to make sure Java would never be used as a serious application language.
MS saw the power of Java to make their platform, windows, irrelevant. They have moved to combat this by simply not supporting Java out of the box. Since most people won't bother to download Java on their own, they know that the chances Java will make any inroads are virtually nil.
Sun saw the power of Java to not only make their expensive, largely proprietary platform Solaris irrelevant; but also saw that Java might "get away", leaving them with nothing. That is, Sun feared that MS or some other company would take over Java by adding more useful base class files and writing a better compiler than Sun. Thus, they would end up being irrelevant both in the operating systems market and in the language they created in the first place.
In my opinion, Sun was justified by fearing Solaris would die (it still will, it'll just take longer), but totally paranoid in fearing they would lose control of Java. Most programmers did not want to use proprietary MS stuff when coding in Java because the more platforms your app runs on, the better. If they wanted to code stuff that would only work on windows, they would not have chosen Java in the first place. Regardless, they made the decision to prevent microsoft from producing Java compilers with proprietary stuff. This was the beginning of the end in terms of Java having any serious impact in the desktop application market, because microsoft refuses to create a Java compiler without the proprietary stuff.
Thus, the promise of Java has not been realized, and may never be. Java based technology is having an impact, but most of that is coming from JSP, a langage for writing web applications. Because it is Java based, will run on any server platform. But because most servers are some flavor of Unix anyway, I don't think this will really change the operating systems market in that sector.
Lets just hope the next company that comes out with an innovative, platform independent language isn't as greedy, shortsighted, monopolistic, and paranoid as Sun. At this point I'd like to see both MS and Sun made irrelevant. As events up to now have pointed out, they are clearly cut from the same cloth.
The first thing you should be considering is what types of applications people are going to need to be running in the labs.
Is there is a requirement that you need to run Windows applications in this lab? Then you really only have one choice: PCs. Mickeysoft does not have a good thin client solution, no matter what they say.
Is the lab is going to be used by "regular" (read: non CS/Engineering) students? The argument again here is strongly for PCs, because the vast majority of users are most familiar with some brand of Windows. Thin clients can be substituted here, but there is a hidden cost: the time required to train people how to use the system goes up considerably. This is due to many more people needing "extra" help to understand how to use the system. These people often need this extra help over and over again because they don't have to use the lab machines that often. That in turn means you have to hire more people to sit around in the labs to help people, and these people are more expensive over time than the computers are.
Now, if the system is going to be used primarly by "techie" type students, such as CS and Engineering, then the actual cost of the system becomes the most important factor. These people already know a thing or two about computers, and those who don't will learn quickly because they'll have constant contact with the lab machines. They'll quickly transition out of the extra help stage and you would only need to hire more help guys for a brief period of time, if you need them at all. These types of students are generally better at solving computer problems, and they're usually pretty good about helping others in an environment where most people have at least half a clue. (They won't always offer help in regular labs due to the high volume of clueless people who are a pain in the ass to give help to, and the "cling" factor: help me, now help me, now me!)
And don't skimp on security. At least have a security wire to keep the machines from walking off. Even dummy terminals usually have monitors that are worth stealing if yours is a piece of crap, or broken. And yeah, I was a sys admin for a university a few years back. I won't go back I tell you! Away, helpless users! Get your tentacles off me!
Slashdot Mirror
I agree that peer "Code Review" is a very important process. We've implemented a code review where I work, and it has been very successful. In fact, I'd recommend using a similar process to developers in almost any situation.
Attendance to our code review is not mandatory (unless you are submitting code), but we really encourage any developer who isn't in "crunch time" on their particular project to attend. Our code review really incorporates all these ideas:
1) Peer review of code going into the testing environment.
2) "Idiot check" testing by peer developers to try and break the code while still in the development environment.
3) A place to showcase new, cool code - code that accomplishes a new goal, or accomplishes an old goal in a significantly better way.
4) A place to describe new processes and/or methodologies that are being put into place for development. (For instance, one of the first things I did when I arrived was spearhead the use of source control software through this process.)
5) A place to fawn over and discuss the latest software and hardware gagets. (mimio started showing up everywhere after one of these)
The only thing that does not go into code review that makes it into the testing environment is "copy n' paste" code - code that is almost identical to another piece that has already been written. (A lot of our ASP pages are like that - call some stored procs, has a form, dropdowns)
I can honestly say that code review has made me a better programmer. First of all, it puts extra sets of eyes on your code - and that makes a difference - I guaruntee people will spot things you did wrong. Secondly, it gives me a chance to see other code that is new and well written. Then I can incorporate those ideas later on in my own code.
Furthermore, it prevents inferior code from going into the user acceptance testing environment and, consequently, the production environment. Every company has a few developers who write second rate or below standard code, and this weeds that code out before it can get in and screw up or slow down the system.
Now, obviously, every developer does not have time to step through the logic on every piece of code that comes through. Thats not really the idea though - thats the developers job. Mostly readability, the use of standard naming conventions, and obvious problems are the sort of thing that is looked for. "Can I debug and/or modify this code if I need to without your help?" is a question I ask myself when reviewing. Generally, the only stuff looked at in heavy detail are pieces that are experimental, bottlenecks, or intricate.
It only takes up about 2 hours of my week, and it makes all the code that goes into our system significantly better, as well as improving our programming skills. All in all, I'd say its a very worthwhile use of our time.
I've done a lot of research into the HTPC (Home Theater PC) market lately. I've read up on XP Media Center quite a bit. I don't think its an ideal solution, but I might buy it if they sold it as a standalone product rather than only with crappy OEM machines. Let me clear some things up for you:
Microsoft disabled the DRM "features" that were originally in the Win XP Media Center. This was done just before it was officially released, and only because of an outcry from the OEMs that were trying to sell these PCs, because those OEMs know that if the recorded videos were all DRMed up, nobody would ever have a reason to choose such a system at approx. $1200 over a $200 Tivo.
The "guide" you refer to is used by just about all the computer TV-tuner solutions out there. It is currently free. How long it will remain so, I cannot say, but it is not controlled by Microsoft; unless they built their own, which is possible - I haven't seen any of these Media Center boxes in person.
Adding HDs is no problem. Make no mistake, this is a regular PC running Windows XP - its just that Windows XP has had an extra layer slapped onto it so that its easier to access all the media functions from a single location.
The major downfall at this point of Media Center in my opinion is twofold:
1) OEM only. Until MS drops this policy, this OS will not become widespread. Most of us geeks who are really jonesing for this aren't willing to shell out $1200 for, esentially, an operating system in a fancy box. Lets see: don't like the case, don't like the power supply, don't like the motherboard, don't like the sound card, needs a bigger hard drive - you get the idea.
2) Proprietary video format. Yeah, currently the Media Center PC only records in a lame ass MS proprietary video format. (Again, haven't seen it in person, so I can't say this for 100% sure, but all I've read indicates this is still true.) I'm sure this will be circumvented, and some haxor will come out with a method of converting it to MPEG 2 or somesuch, but it still sucks.
So, basically, you'd want to buy one of these things under the following conditions:
1) You aren't a *nix zealot. (Unless you are a hypocritical *nix zealot)
2) You're in the market for a new PC.
3) You don't care about the specific parts in said PC.
4) You don't care about anyone using a non-MS operating system being able to play recorded files you send them.
5) You don't care that you will now have a computer in your living room, and all the ugliness/noisiness that entails.
The latest issue of PC World hit the nail on the head, when they said (parphrasing): "this is really only attractive to people living in either a tiny apartment or a college dorm." In other words, people who already have to deal with a computer being in their living room in the first place.
You didn't happen to WTFA did you?
All kidding aside, its obvious to me that at least one of the following must be true:
1) This paper was written purposely to fool an ordinary reader (such as a journalist) into believing that human influence extends further than it really does.
2) The journalists who wrote the headlines and articles describing this paper clearly intended to fool readers into believing that the paper describes human influence accurately, even though it does not.
3) Journalists are idiots who don't know their head from their ass, and are incapable of descibing a paper such as this accurately.
I say this because the headlines and articles all describe this paper with inaccuracy, and the fault must lie somewhere. Most people posting here obviously believe its option #1. Its obvious to me that you either believe that journalists are either environmentally biased or complete morons (Options #2 and #3, respectively).
Frankly, I believe all 3 options are true to some extent.
The other gamers in the place looted his corpse.
d00der: "Fake ID +5 to fool police? This guy sucks. I already have a Authentic Looking Fake ID +9!"
3l337H4X0R: "Is there a blacksmith close? I need to repair these shoes before I sell them."
MrKim224: "Stop stealing my stuff I'm only unconscious! Wait until I respawn, I'll get you bastards!!"
SYSTEM: No one hears you. You are unconscious.
Not sure why this keeps getting modded up in every single thread..
Moderators, I think its time we declared this the new All Your Base/Hot Grits/Natalie Portman/etc "Slashdot catch phrase" (TM). It should be modded down as such. After all, its only funny the first couple hundred times.
Please check the following link for information on how Slashdot catch phrases (TM) and other Memes work: http://totl.net/Cool/
You may now continue your regularly scheduled productivity draining activity.
The slashdot ID.
user: slashdot223
pass: slashdot
Thats three Blizzard years(TM). Expect to see Starcraft 2 in 9 "regular years".
Take a look at this website, it'll tell you the current locations and distances of spacecraft on solar system escape trajectories. (Pioneer 10, Pioneer 11, Voyager 1, Voyager 2)
http://www.heavens-above.com/solar-escape.asp
Pioneer 11 is actually the least distant of all these spacecraft - and is also essentially a manmade rock at this point - its no longer operational.
Voyager 1 owns the record for farthest out, and will for the forseeable future, as it is travelling considerably faster than the others relative to the sun.
Also interesting is that the Voyager twins will be the two farthest spacecraft out once Voyager 2 passes Pioneer 10.
This should be in late November 2021 by my admittedly crude calculations. Take Pioneer 10's 13.986 AU lead, divide it by the 0.727 AU V 2 is gaining per year on P 10, and you get approx. 19.24 years. Add that to todays date and you arrive somewhere in late Nov. 2021 - I'm not going to bother with the exact date.
Commander Tuvoc (Voyager) bled green in several episodes. As far as I know blood color continuity has been consistent across all Star Treks.
Enterprise has not dealt with alien blood colors to this point to my knowledge.
Of course, this'll probaly cost me Karma as its pretty much offtopic.
grrrrr.
Would someone please reply with the slashdot username and password for NY Times?
Thanx...
I'd like to make some comments from the point of the research I've done so far on this subject, in relation to some of the things you've said.
First, on hard drive failure:
Hard drives fail. This seems obvious to me, I was a network admin in college and I saw a lot of hard drives fail. Pity the poor fool who tries to transport them.. click click, click click.
My Tip: Burn. CD burners are cheap, reliable, and the media is super cheap. Yeah, an entire movie won't fit on on CD. But you won't be playing from the CD, the idea is merely to have a copy on permanent media in case of HD failure - then put it back on the new HD when you get it. So, make sure you have your media box hooked to your LAN and copy those files to your CD burning machine. No burner? Buy one. - Note: My Pine drive shipped with Nero full version, thus I recommend buying one, because otherwise I would have bought it seperately anyway.
Noise:
Computers are noisy. This is probably the biggest hurdle to overcome in a entertainment center pc.
My Tip: abandon convention. To combat noise, I'm looking at using a specialized watercooled case. I realize this case doesn't fit the traditional entertainment device size/configuration - but I've abandoned that notion and just committed myself to buying the quietest setup I can find and figuring out a way set up my entertainment center around it so that it doesn't look out of place. I have to admit, at first, a "dumb terminal" type setup seems like a good idea, but most of the stuff has to run from the client, not the server. You basically aren't buying yourself any help because your powerful video card and processor have to be on the client anyway, and those cause heat - so even a dumb terminal is going to be noisy.
TIVO Functionality:
A must-have. Clearly an entertainment PC is only going to be more than a fancy looking DVD player if you include video capture and playback.
My Tip: ATI All-in-Wonder. The ATI All-in-Wonder is still the best video capture card on the market. Note that it comes with a remote. NVidia has one now, but reviews are not as glowing. Program guides for both are free.
Output:
TV resolutions aren't very good. My research suggests that even new, fancypants TVs resolution isn't good enough to use for a computer.
My Idea (not advice): projector. Currently I'm considering buying a high resolution computer projector for about the same cost as a very nice TV (a.k.a $2,000-$3,000). I've seen one of these at one of the places I've worked, and I was impressed. It showed both a 1024x768 computer screen and DVD playback very well. (Comon, I *HAD* to play a DVD on it!) The only issue here is obviously light - projection does not work well in a room with a lot of light. Even though otherwise this is an ideal solution (huge screen, relatively low cost, high res) - I'm not sure I can live with the tradeoff. I imagine for normal TV viewing I will still go with the regular TV I already own. My super-sneaky idea for a "close" monitor - build a flap to come up closer to the projector. I doubt the text would be readable, though. I'm still mulling over the possibilities.
Alright, maybe the guide does suck. Why not list:
A) Kits you think are better
or
B) What parts you would use that are better
or both.
Its easy to say "this sucks" and leave it at that. The rest of us would like to know, if there are better options, what are they? The whole point of the kit is to make this sort of thing accessible to geeks like me who want to do a watercooling solution, but don't have the time to learn everything there is to know about pumps, tubing, etc - and don't want to spend the additional time trying to locate the best parts for the job. What we need is a list of parts, where to get them, and if there are tricky parts to the job, a short guide. I'd settle for a list of the best parts suited for this task though.
"no bets on when the technology covered will really be available though"
Didn't you know? Its five years away, of course.
Just like everything else...
A case with a lock on it? - You'll have to crack into it without a metal cutting saw.. and thats gonna be noisy and take some time. (Not to mention leave a mark.) Note also that most manufacturers sell cases with BIOS level intrusion detection built in if you want it.
I applaud the efforts of junior MacGyvers, but if you really want to be secure, there are obviously better solutions.
Speaking of computer security, that reminds me of the time when the CS department at the University I went to got a bunch of brand new lab machines. They all had intrusion detection, which we CS dept. admins thought was pretty cool. We told the campuswide IT guys that we needed them secured in place. They dragged their feet on it. A month later, the CS department threw a Lan party in the same building (though not in the same room) and some enterprising students used it to cover the theft of 4 of the new lab machines. Security wire and cameras were in the room within a week. As far as I know the stolen computers were never recovered. We took small solace in the knowledge that the computer beeps at them and displays a brief annoyance message now before booting the OS. That is, as long as they opened the case and didn't flash the BIOS.
In psuedo code, I might agree with you. In a specific language, requiring it compile, no. Yes, a good programmer should know basic concepts, but more importantly, a good programmer should be able to use basic concepts in a language independant manner. Why would anyone want a programmer to know how to code a linked list? Its a waste of time. No one is gonna code it from scratch, its a trivial and demeaning question. I know if I went into an interview and someone asked me for code like that, I'd tell them straight up that I'd look that up in a book or online, and that it'd be a waste of my time to write it. Maybe if you're hiring greenhorns straight out of school you might ask a dumb question like this just to make sure they aren't completely clueless, but for experienced programmers its frankly a poor test to find a good problem solver. Really, in my opinion, it shows a lazy interviewer who hasn't bothered to come up with good questions.
I'd argue that you are not only eliminating the worst candidates with that type of test, but also the best. The best programmers do not remember code in a specific language. The best programmers do not waste valuable "brain space" remembering trivial crap like linked lists that none of us would ever write in practice (since its been written a bazillion times anyway).
Writing code is not done in a sterile, non-networked environment without books. We do not code on the top of Mount Everest. Good programmers certainly will not waste a companies time and money by writing a linked list from scratch. Knowing how to write code in a specific language is one thing, knowing how to write some trite piece of code that you haven't written for 5, maybe 10 years is another.
I say, if you're going to ask for code, ask the interviewee beforehand what types of things they've been coding recently and ask a code question along those lines. Do a little work yourself instead of asking some lame ass question that no one would have any good reason to remember the answer to. Hope you like those greenhorns or anal rote memorizers you've been hiring. They must be either close enough to having been in school or a one language pony to remember useless junk like exactly how to write a linked list.
Frankly, I'd be more apt to hire a candidate who told me writing a linked list was a stupid question and that he'd just look up sample code than one who wrote it perfectly. I don't want droids working with me, I want problem solvers. That is what programming is all about, after all.
In short, unless you're hiring for the position of linked list expert extrodinare, asking for on the spot compilable code for a linked list is most likely a waste of time for the interviewer and the interviewee.
I have to wonder, was exclude all stories by author implemented solely because of Katz? If it predates Katz, certainly it was a prescient decision.
At the very least, I would be willing to bet that half of all slashdot user accounts were created to eliminate Katz stories from showing up on the main page. I know mine was. Maybe this should be the next poll?
Did you create your slashdot account so that you could eliminate Katz stories?
*Yes
*No
*I don't have a slashdot account you insolent clod
*I just use cowboyneals account
And what are the percentages on which authors are excluded by slash user accounts? I'll bet Katz is winning by a landslide. 70%? 80%? More?
On a side note, you might wonder why I'm replying to a Katz article if I exclude his stories. Different computer, wasn't logged into slash as me.
Oh, also, I wanted to add a me too section. Quite a number of posts have pointed out an earlier slash article disagrees with Katz about the magic of myth bullcrap and points out the true origins of Star Wars in sci fi pulp fiction. To those posts I say: Me too!
He's probably not a card-carrying, slashdot reading member of the lumpenintelligentsia such as yourself though, so you really don't care.
/.
I do care, thats why I my post tries to point out that the target of this trap is often poor people who often feel they have no alternative but to steal. Seems you're the one who doesn't care:
Seems now the refrain goes: "Give me your tired, your poor..." - so I can throw them in jail.
Only if they're car thieves.
I'm not even going to comment on your irrelevant stuff about Abscam. A program busting rich congressmen for taking bribes and one busting poor people for stealing cars aren't even similar in the terms of my argument.
Take some advanced college courses in criminal justice or sociology, then you might actually learn something and not spout mindless crap like this on
I did learn something in those classes. This isn't mindless crap, but I'd argue your double talking "you don't care about them, throw the bastards in jail" post is.
Who cares? You SHOULD care. The police are busy wasting your tax dollars and mine in order to put more poor people in jail. Meanwhile, corprate American criminals continue to rape and pillage the country while you and I pay for it. What I'm saying is, shouldn't we be spending money to trap corporate theives costing us billions than poor people costing us millions? Shouldn't we be after people who steal out of greed rather than those who steal out of need?
I haven't watched the X-Files for at least two seasons, after the tremendous sucking of three seasons ago. The fact that anyone is still watching that show astounds me.
Pulling this sort of crap with the best characters the show had come up with for years doesn't surprise me.
The show really should have ended with the movie, it was showing signs of sucking then. Afterwards, the show took a dive into a soap opera set in a parinoid delusion of space alien invasion. Yeah, I know the show always had that bent, but it went off the deep end after the movie. We begged them, go back to the monster of the week! But alas, they didn't listen, and alienated (literally) many of their fans. All the mystery was removed from the mysterious characters, at least, those that were still alive after the movie went on an important character killing binge. With the skepticism removed from the Sculley character, the show had no way to ground itself in some semblance of reality. Really, there was no one and nothing left for the show to run on.
I find it merciful that the show only has 4 episodes to go. X-Files, you once were a great show, but you went on far beyond your prime, and you've tarnished your image in front of many of your fans.
The preceding opinions are my own and do not neccessarily reflect those of geeks whose eyes towards the X-Files glazed over long ago.
Poor people do not own cars worth stealing.
Because they are poor.
How many car alarms did you ignore today? 1? 10? More?
Lets examine the situation here:
A "late model sedan" a.k.a late 90s Honda is put in a poor, run down neighborhood and left with the doors unlocked.
Poor people recognize this car:
A) Does not belong there. (Despite what the article says)
B) Is not owned by anyone they know.
C) Is unlocked. (By trying the door)
For a person walking by this car represents easy money. If you were poor, would you walk by $10,000+ sitting on the street?
Granted, most of the people who steal this car are going to already be car theives. But you have to admit there is a distinct possibility that this car could tempt someone who was not a car thief to become one.
Seems now the refrain goes: "Give me your tired, your poor..." - so I can throw them in jail.
Take an introductory college course in Criminal Justice or Sociology. It'll really open your eyes.
I'll leave you with this:
Who steals more: poor people scraping a living, or rich people in the boardrooms of corporate America? What theft has more impact on the lives of most people?
On an oddly related point, I saw a thing on TV recently about new inventions and some contest run every year to give an award to the most promising new invention. This one wasn't voted most promising, but it was cool:
One guy invented a speech feedback system that picked up your voice in a headset microphone, changed it slightly, then fed it back into earphones you had on. He previously had a serious studdering problem, but this device allowed him to speak almost perfectly after having used it repeatedly to practice speaking. He demonstrated that the device could also be used to make a person speak authoritatively, cheery, fast, slow, and various other ways as well by changing the feedback that they were hearing.
Pretty amazing, how the brain would change the voice to match what it thought was normal almost immediately. People wouldn't have any problem adjusting voice inflection to match the device assuming it provided feedback.
I'll try and give a few reasons for the eventual demise of Solaris in the following long winded spiel:
First off, I don't subscribe to the imminent demise of Unix conspiracy theory. Various flavors of Unix have a strong hold on the server market, and with good reason. I don't see this changing. What I do see changing is the market share of Suns Solaris platform.
Ever tried to install and run Solaris on PC hardware? Trust me, its painful. Now, I know what your going to say: a lot of servers don't run on PC hardware, so having a working install and running well on such hardware is irrelevant. Thats where I might begin to disagree with you.
In the next 5-10 years, I expect to see a serious movement towards _networked_ distributed computing in the server market. I know this is in use in some applications now, but I think we've only seen the tip of the iceberg. Distributed power + useful real world application (aka not Seti) = market bonanza. And with computing distributed among many machines in a network environment, I expect to see a movement gravitating towards the use of cheaper PC hardware over massive proprietary server machines.
Think about it: what sort of things are most servers doing now? Serving data, serving web pages, handling user accounts, backing up data, and maybe running some programs of some kind. Now, you're probably saying, its that last part that the big dedicated servers are all about. True, but consider this, what if each of the programs were running on its own PC server? Wouldn't that be the same as running on a small portion of a large servers CPU cycles? And if its not, then we're probably mostly out of Solaris's market anyway. Do we need that massive server if we can simply make a lot of little servers look like a big server to anyone who is using it? There is potential here to make data backups mostly irrelevant, since each PC server holds an overlapping piece of the total data, the loss of one or even a few machines doesn't matter. The emergence of IDE RAID makes this an even more distinct possiblity, because of protection at both the machine and the network levels. There is potential to scale properly and closely to a clients exact needs, causing a client to spend a lot less money on a computing platform. Finally, there is potential for maintainence costs go down, due to the lack of needing someone with special skills to handle specialized hardware. There are disadvantages, I won't deny that. However, I think when it is properly done, the advantages of this type of server environment will vastly outweight the disadvantages.
And as for that imminent demise of Unix thing, I think the future is rather bright, because I think it will be some flavor of Unix that brings this sort of technology to the masses. If I'm right, that flavor of Unix will not be Solaris.
Please, there weren't enough obscure acronyms in this piece for your average government agency. We need more. Have someone get on this ASAP. We need more acronyms PDQ. We need to assign a DOAP and make sure they PATFT. That way we'll all be MHIB. Clearly LIAP for posts or we'd have LODT. So lets KUTGW with OAP and we'll all be VAFWWH.
I consider myself a tech-head, and if I can't make sense of a tech article at a glance after getting a Bachelors in Computer Science, something is wrong. I don't even know if I'm interested in this article. It has something to do with the internet, emergencies, and 9/11; and the rest is friggen jibberish. To add insult to injury, michael the slashdot moderator adds an unrecognizable acronym of his own!
PSTN? GETS? IEPREP? Not to mention the slightly better known RFC and IETF? This is crazy. IMHO, I shouldn't have to follow a link just to find out WTF the article is about. These kind of posters need to STFU or slashdot will be a FUBAR POS that just wastes my time.
DOAP: Designated Obscure Acronym Poster
PATFT: Post All The Friggen Time
MHIB: Much Happier I Bet
LIAP: Length Is A Priority
LODT: Lots Of Descriptive Terms
KUTGW: Keep Up The Good Work
OAP: Obscure Acronym Posts
VAFWWH: Very Appreciative For What We Had
That was the idea, when Java first arrived on the scene..
Since Java was platform independent, you could write a Java application, and it would run the same no matter what was running underneath, be it Windows, Linux, or what have you. This meant that if a lot of truly useful software had been written in Java, Linux would start to become a seriously viable alternative, because it could run the the truly useful software just as easily as windows. As the first poster in this thread astutely observed, MS and Sun combined their powers to make sure Java would never be used as a serious application language.
MS saw the power of Java to make their platform, windows, irrelevant. They have moved to combat this by simply not supporting Java out of the box. Since most people won't bother to download Java on their own, they know that the chances Java will make any inroads are virtually nil.
Sun saw the power of Java to not only make their expensive, largely proprietary platform Solaris irrelevant; but also saw that Java might "get away", leaving them with nothing. That is, Sun feared that MS or some other company would take over Java by adding more useful base class files and writing a better compiler than Sun. Thus, they would end up being irrelevant both in the operating systems market and in the language they created in the first place.
In my opinion, Sun was justified by fearing Solaris would die (it still will, it'll just take longer), but totally paranoid in fearing they would lose control of Java. Most programmers did not want to use proprietary MS stuff when coding in Java because the more platforms your app runs on, the better. If they wanted to code stuff that would only work on windows, they would not have chosen Java in the first place. Regardless, they made the decision to prevent microsoft from producing Java compilers with proprietary stuff. This was the beginning of the end in terms of Java having any serious impact in the desktop application market, because microsoft refuses to create a Java compiler without the proprietary stuff.
Thus, the promise of Java has not been realized, and may never be. Java based technology is having an impact, but most of that is coming from JSP, a langage for writing web applications. Because it is Java based, will run on any server platform. But because most servers are some flavor of Unix anyway, I don't think this will really change the operating systems market in that sector.
Lets just hope the next company that comes out with an innovative, platform independent language isn't as greedy, shortsighted, monopolistic, and paranoid as Sun. At this point I'd like to see both MS and Sun made irrelevant. As events up to now have pointed out, they are clearly cut from the same cloth.
The first thing you should be considering is what types of applications people are going to need to be running in the labs.
Is there is a requirement that you need to run Windows applications in this lab? Then you really only have one choice: PCs. Mickeysoft does not have a good thin client solution, no matter what they say.
Is the lab is going to be used by "regular" (read: non CS/Engineering) students? The argument again here is strongly for PCs, because the vast majority of users are most familiar with some brand of Windows. Thin clients can be substituted here, but there is a hidden cost: the time required to train people how to use the system goes up considerably. This is due to many more people needing "extra" help to understand how to use the system. These people often need this extra help over and over again because they don't have to use the lab machines that often. That in turn means you have to hire more people to sit around in the labs to help people, and these people are more expensive over time than the computers are.
Now, if the system is going to be used primarly by "techie" type students, such as CS and Engineering, then the actual cost of the system becomes the most important factor. These people already know a thing or two about computers, and those who don't will learn quickly because they'll have constant contact with the lab machines. They'll quickly transition out of the extra help stage and you would only need to hire more help guys for a brief period of time, if you need them at all. These types of students are generally better at solving computer problems, and they're usually pretty good about helping others in an environment where most people have at least half a clue. (They won't always offer help in regular labs due to the high volume of clueless people who are a pain in the ass to give help to, and the "cling" factor: help me, now help me, now me!)
And don't skimp on security. At least have a security wire to keep the machines from walking off. Even dummy terminals usually have monitors that are worth stealing if yours is a piece of crap, or broken. And yeah, I was a sys admin for a university a few years back. I won't go back I tell you! Away, helpless users! Get your tentacles off me!