I took the Core+OS version. From what I understand, they redid the structure after I took mine? Hopefully it's less of a joke now. Mine didn't have any technologies newer than USB 1.0 and Win2k.
Distinguish CCNA from A+
on
Network Warrior
·
· Score: 5, Informative
Having just obtained both CCNA and A+ certifications within the last 9 months, even though neither was really necessary for my current job, I feel the need to defend the CCNA exam by contrasting it with the A+ exam.
The A+ exam is a complete joke. It might have been relevant 10 years ago, but isn't anymore. The vast majority of questions were completely irrelevant today, and mostly irrelevant for any computer faster than 500MHz. Most of it was pure memorization, the kind all those cram books are for. There were very few real-world questions -- less than a half-dozen. There were even a few questions that were nonsense -- none of the answers were applicable, much less valid, and in some cases the question didn't even make sense.
The CCNA, however, is a killer. I took mine in May 2007. The first part of the exam is roughly one-third memorization questions, one-third diagram interpretation questions, and one-third real-world questions. Most of the diagram questions are trick questions with multiple realistic-looking answers. (In other words, you can't just look for the "obvious" answer.) This part of the exam is meant to test your grasp of networking concepts.
The second part of the CCNA exam is what really gets you, though. It's all about configuration. Most of it is spent in a simulator. And not just a simulator for one router or switch, but a simulator for an entire network. One of my questions involved configuring 4 different routers and 3 switches. Oh, and they can disable parts of the simulator to make your job harder -- like having to diagnose a connectivity problem without being able to ping or traceroute. And yeah, they like to throw multiple IOS versions at you to make sure you know the different variations of the commands (especially for switches). You cannot cram for these simulator-type problems.
I tend to consider myself a pretty smart guy. I've been working as a network admin for 10+ years, albeit not with Cisco equipment. I aced the classwork for the CCNA courses without putting forth any effort whatsoever. I did homework in class and never had to come in after hours to catch up. And yet, I had ~45 seconds left on the timer when I finished part 2 of the CCNA exam. It's that tough, and they've got it timed down to the last minute. You do not have time to flounder and guess.
If you don't know your stuff backwards and forwards, you are not going to pass the second half of the CCNA exam. It's that simple.
Now, having said all of that, remember: the CCNA is the entry-level exam. It's not meant to certify that you can walk into a company and rewire an international infrastructure by hand. It's meant to certify that if you put me in front of a router or a switch or a small network that is having problems, I can most likely figure out what the problem is. The building-huge-networks stuff is part of the CCNP, not the CCNA. (The first CCNP class is, after all, "Building Scalable Networks".)
I see plenty of haterade about the CCNA exam, but I never seem to see it from people that have taken the tests. And I have to wonder: for all of those exam-crammers with CCNAs that everyone seems to know, when did they get their certs and are they current? I doubt it.
Disclaimer: I'm the guy that wrote TorrentSpy (the application, not the web site) and have contributed a small amount of code to the Python/core BT client and tracker. I haven't written any code for BT in a while, nor have I chatted with Bram in literally years, but...
From what I remember, Bram always viewed the Python/core as a sort of "reference implementation" -- it was never his goal to make the Python client or tracker the end-all be-all.
Why is it then surprising that he'd want to bring on a client that doesn't have to be Open Source, and thus doesn't have to be clean and perfect, but is still sexy as hell? He still keeps his reference implementation that supports the features and is easy to reimplement a dozen times in two dozen languages... but he also gets something he can brandish at anyone who wants to throw money at BT.
Remember that his goal since incorporating has been to legitimize and broaden the adoption of BT. A sexy client is a huge step towards that goal. It's not like Sony or the MPAA or whomever is going to distribute a customized version of Azureus any time soon -- it's a beast! But a custom version of uTorrent? A 1MB executable that you could throw on a CD that requires zero install? YTF not? Remember also that Strigeus has been working towards licensing out the core engine for uTorrent.
Plate. Shrimp. Plate of Shrimp.
Some of you are excessively paranoid. You know that, right?
I have, right next to me on my desk, a stack of 1x CDRs that I burned back in 1996 on my double-height SCSI 1x CD burner. There are several different brands, but the majority of them are Maxell and Verbatim.
Each and every one of them still works perfectly. And no, I have not stored them vertically in a cool, dry place. Some of them were left on the front seat of my car, under the seats in my car, and various other random places.
On these CDs I have backups of web sites I used to work on, horrifically low-res quicktime video clips of Babylon 5, and a ton of MP3s. Yes, MP3s from 1996. No ID3 tags, and in ALL_UPPERCASE because of the Joliet/ISO FS wars from the time, but still perfectly listenable. Many of them ripped and converted on mindbogglingly slow 200MHz machines before MMX, and many of them downloaded over HTTP from web sites back before P2P was even thought of.
So, yeah, the article is bunk. My stack of nine-and-a-half year-old CDRs is working just fine, thank you.
This is a totally shameless self-promotion, but...
I recently completed the first half of a book that features a program that does exactly this: plugs into your software and calcuates trust networks, then takes it one step further and applies firewall filters based on those networks. Basically, a P2P version of PeerGuardian, but smarter and more plugin-friendly. The author is Xochitl Green, a web programmer who quickly discovers that she's out of her league because there are people out there that don't necessarily want self-securing computers.
The first chapter of "Trust Network" is available for download from my site, and you can get the first half (up to chapter 18, about 150 pages) through lulu.com in either PDF or printed form, for what I think is pretty cheap. (There's even a large-print version for the visually-impaired.)
I'm pushing for the second half to be out by March 30. I don't claim to be Gibson, Gaiman, or Stephenson, but it's not Star Trek fanfic, either.
While I agree with the point you're trying to make, you're basing it on some assumptions that are not valid. First, dad isn't just a software geek, he was an ATC for 20+ years in the US Navy with a couple of years as an ATC in the private sector thrown in. He's used everything from flight strips and binoculars to the latest and greatest software. He's also been using computers at least that long and knows his way around every version of DOS and Windows and is even really getting into *nix. Hence why he is in the position he is in: he knows both sides of the game. But I know you weren't on the attack, and I'm not trying to be on defense, just clarifying a bit.
Secondly, the ATC folks definitely know what they are talking about. Hence why the software team is made up of a bunch of ATCs overseeing all the programmers and database weenies. This isn't some outsourcing gig we're talking about here. It's people redesigning their own world from the inside.
The point I was trying to make was not "throw out the old crap and make everything whiz-bang". Nor was I trying to paint the world over with Apple's HIG. I was attempting to point out that the system that is in place has become so esoteric and counterintuitive that it is actually hindering the training of this generation of ATCs. Or, like the topic says, "know your userbase".
When people have to make a conscious effort to rewrap their brain around some foreign thing every time they go to work or come home, it becomes stressful. Stress leads to mistakes. If you speak only Chinese at work but only use American Sign Language at home, don't you think you are going to slip up every now and then? Potential new ATCs have been raised on the same basic interface rules for the past 20+ years. Old ATCs have been trained on an orthogonal set of interfaces for at least as long.
The problem isn't that the interface that the old ATCs use isn't valid. Obviously it is valid, as it has worked for so long. The problem is that the interface they are used to (knobs, switches, dials, paper flight strips, etc) doesn't translate well to a computerized version, and even if you shoehorn it in, that computerized version doesn't feel right to younger kids, who are the whole reason for making the translation in the first place.
Remember how much controversy was stirred up when Apple made the first pretty version of Quicktime for Windows? It had jog dials and a draggable "drawer" that just didn't make sense to most people. Sure, those interfaces are fine for 3D-space, but for half of the people the digital versions suck. (Notice how both of those mis-features have been retooled since then?)
You mentioned that efficiency is the crucial element, training be damned. I agree. But I think you are missing the fact that the old ATCs are efficient not necessarily becuase the system is good or bad, but because they've been trained to be efficient at it. Potential new ATCs have been trained to be efficient at using current HIG. Why retrain them?
So what's the right answer? Build a system with a whiz-bang interface that any XP user off the streeet can figure out, but is nothing like what has been in place for the last century? Or build a system that people who have been ATCing for years can relate to, but is counterintuitive for everyone else and requires extensive mental conditioning to use? Or build a system that tries to mesh both paradigms but is still nonintuitive to both new and old ATCs? I certainly don't know.
My dad is helping the FAA and the US military design and roll out the next gen ATC software here in the US. He comes home and tells stories that make my skin crawl.
The first version of the software was built using standard current interface guidelines and widgets and the testing group that had no experience with older ATC systems were wowed at how simple and yet powerful it was. Pretty much any random person off the street could look at the screen and easily figure out what was going on and how to do various basic tasks. When that version was demoed to the ATC union the union freaked out at how different it was and thus began a cycle of making it more and more backwards.
So, nowadays the next gen ATC software almost exactly replicates the UI of the old non-computerized and semi-computerized systems. On-screen toggle switches and dials, that sort of thing. The FAA and the ATC union have decided that retraining all of their ATCs to use modern computer interfaces would be a Bad Thing. When the computer screen doesn't exactly replicate the interface of the 50+-year-old systems, they freak out and scream bloody murder. On the flip side, kids coming into the field today that have been using computers most of their lives are finding the interface to be counterintuitive to the point of being almost unusable. Middle-aged workers who are both highly proficient ATCs and home computer users report that switching between the two types of interfaces each night when they go home requires conscious effort on their part, since they are so orthogonal.
So who wins? Historical inertia, of course. Why fix the problem today when you can wait for your successors to fix it in 25 years?
Yep. I just last night unwrapped my brand-spankin'-new copy of Splinter Cell: Pandora Tomorrow. I did the install thing (2+ GB), after which their little "video card compatibility tester app runs". Lo and behold, my card wasn't compatible due to inadequate pixel shaders. Now, the app doesn't stop your from running the game, it just throws up a little red "X" that is very easy to skip past. (Especially given that on most video card compatibility apps, that just means "you won't be able to see every little facial hair at 300fps". Oh darn.) I didn't think much of it at the time, and I just continued on to run the game. No joy. It just blinks out without any sort of dialog box or anything. I fiddled with the resolution and detail settings, even mucking about in the INIs, for a good 30 minutes, still no joy or any error at all. Just a resolution flicker, a half second of sound, and back into Explorer.
Went to the web site and found out that my video card is specifically excepted from the compatibility list. I have a GeForce 4 MX 440 with 128MB of RAM. This card is what, 2 years old? It's certainly not a low-end card, to be sure. And, given the above metrics, the MX series is a good 25% of the user base. And yet, it doesn't have the all-wonderful pixel shaders and is somehow obsolete? I tried using 3D Analyzer to disable the pixel shader requirement for the game, but still no joy. But at least then I got an actual GPF instead of nothing!
But wait, here's the best part -- the SC:PT demo runs without a hitch on this card. That's a little strange, eh? Sure, I can understand that maybe the pixels shaders aren't needed until some end-level of the game that the demo doesn't have. But doesn't that imply that maybe the pixel shaders aren't really all that necessary to play the game? What ever happened to a nice graceful degradation of features? If I play Homeworld2 I can disable graphics features until I'm looking at fighting polygons. (Not that I have to, mind you, as HW2 worked out of the box on my system.)
Why not upgrade, you ask? Why should I? My card is still good enough for every other recent game I've tried to play. I'm not the type of gamer that needs to see individual strands of facial har whizzing by at 300 fps. I'm not saying we should go back to 2D sprites, but this is just ridiculous. If I were a tinfoil-hat-type, I'd be ranting about a nefarious plot by the gpu manufacturers.
So I gues my issues are thus:
How is a 2 year old card with 128MB of RAM already considered obsolete by a game developer?
Why can I play the demo but not the full game?
How can someone, in good conscience, release a game to a market that still has a boatload of these video cards (25%!!!) if the game doesn't work on them?
How can a game developer, in good conscience, release a game that doesn't at least try to degrade gracefully?
WTF is up with running the video card tester after installing 3GB of game onto my machine?
You know, I've read the articles about how hard it is to get a decent game published. (Probably not in this case, given the publishers, but still.) I try to support the developers by buying a legit copy. I'm staring at a shelf full of game boxes. And yet, this kind of crap happens. While this isn't going to turn me off from buying games altogether, I doubt I'll be paying for any more Ubisoft games in the years to come.
SupaDupaSpy Syd and Noah Hicks (Peter Berg) are on a plane back from Madrid (?) and he asks her why she never met up with him in Rio (?). She's shocked and has no idea what he's talking about. He'd encoded the meeting invite in the headers of a forged spam email. She never got it because she has her computer automatically delete all spam.
Pretty sad when/. is scooped by the pretend-CIA by over a year.
Just to be the voice of dissention... I'm apparently the one person in the world for whom the PHP+Apache2 combo doesn't actually work right. Yes, it mostly works, but I'm constantly having to close out the windows that pop up when a thread goes boom. Yes, I am running it on XP (with all of the necessary service packs and hotfixes), but that shouldn't invalidate the fact that PHP+Apache2 isn't production-quality stable. And yes, I am 100% sure it is PHP causing the problem, as the errors only started when I started using PHP, and they increase as I convert more and more of my site to PHP.
Now, if I'm getting all these errors, why am I using it? Because it's still better than the alternative. When any DLL goes pop under IIS you get really flaky and esoteric things that start happening. Under Apache2 it just nukes a thread, which Apache2 diligently respawns and goes on with life. I can deal with clicking OK on a dozen windows a day if it means I don't ever have to worry about restarting my web service.
PHP+Apache2 isn't perfect, but it's good enough for what I need.
My review from IBList.org
on
Singularity Sky
·
· Score: 2, Interesting
I reviewed this book on IBList.org last year. I figure I'll repost it here just to add another voice to the chorus.
-----
Economics, espionage, nanotechnology, black holes, social enginerring, and carnival phenomena. This book winds all these disparate subjects together into one (mostly) cohesive plot. This tends to lead to parts of the book reading more like college textbook excerpts than light sci-fi reading, but that may very well increase the appeal for the hard-core geek readers. The pacing occasionally suffers from the massive amounts of technobabbling exposition, but you still slog through it like a rubbernecker watching a car crash -- you just can't wait to see what it all means.
The character development is better than average, though there could have been more character-building scenes without significantly slowing the pace. Indeed, the technologies and concepts often get more ink than the characters do. (Because, really, there's only so often you can be hit over the head with the "socialism/marxism/communism/*-ism is bad!" bat before you're ready to start skimming instead of really reading.)
Overall, this was a good book. It could even make a good series, should Stross continue to write for it. College students pumped up on technobabble and economic/social theory will breeze through it, but the rest of us will still enjoy it.
The tracker already has all of this data. It has a count of how many people have completed as well as how many people are incomplete and how much they have left to go. (Like this, a CGI script that just reads the tracker's state file.) In fact, there's even a stats generator that makes pretty graphs out of it all. We're working on getting the UI elements to display these stats integrated into the core (example), but for now there are patches that do it. (Shameless plug.) To top it all off, the tracker generates log files that are plain old HTTP logs, and are parseable by pretty much any web stats generator.
If AOLTW/WB/etc ran their own tracker they'd have all the stats they needed. If you upload a torrent to a 3rd-party site and are the initial seed and want to see the stats, pretty much all of them nowadays let you at least see how many total downloads there have been. It's not an issue.
Another poster already explained some of the features you had questions about. As for the implied "why would I go through the extra step of having to configure the Launch button or use the Check button?" question... the default BT3.x client does not as yet support the ability to check the current number of seeds and leeches (people with 100% of the file, and people with something less than 100% of the file), nor does it tell you which sub-files are complete/incomplete. Checking the seed/leech count is a quick HTTP operation so it is done by default when you open the app. Checking the files for completion, however, can take quite a long time for large files, and if it is a single-file torrent then you really don't get much more information than you would with the normal BT client. Therefore, it's a manual operation.
The way TSpy is intended to be used and is actually most commonly used is as a quick way to check the seed/leech count. This is because for most of the file trading that is currently done with BT, a torrent will stay active for a week at most and then go dead with no seeders. It sucks to launch the BT client, have it check your 2GB torrent, then wait an hour only to get stuck at 60%. TSpy can tell you in 5 seconds the odds of you getting a good download. The other usage that has become quite popular is for a seeder to pop up TSpy every now and then to see if anyone is still leeching. If not, they can take down their BT client and go seed something else. In both of these cases you don't necessarily want to launch the BT client, which is why that is also a manual operation.
The more granular file check is in place to help people recover something if it looks like their torrent is dead and they don't yet have all of it. It'll tell you which files are complete and how much you are still missing. And, before BT3.2 added in the download count counter, it was amusing to just keep rechecking the files to watch the bits pour in.
Admittedly, the extremely casual BT user probably won't have much use for TSpy. They'll probably be getting only the most popular and long-lived torrents that are almost guaranteed to always have seeders. But for the people that are trying to keep track of files that start and die out within a few days' time, TSpy can be rather useful.
Yeah, it's buried way down in the body of the article, but it's still there. w00t! I'm the guy that wrote TorrentSpy. The web site that the link points to is running on my PC at home on cable 'net access. 300 hits in 3 hours. Not too bad so far, I guess. My gf is gonna kill me if I get slashdotted, tho. There are worse ways to go, I guess...
I've had several people express interest in an *nix version of TSpy. It's written in Delphi, which theoretically means it shouldn't be too hard to port over to Kylix. If anyone has Kylix experience and would like to help out, feel free to contact me at the email on my web site.
I'm gonna go tail -f my log file and listen to my network bog down.
I think it is kind of funny for millions of people to live on a sand bar that gets hit over and over again by hurricanes
Yeah, but when you can walk out of your house in the morning and be standing on a sandy beach with the sun shining and warm water 9 months out of the year (or all-year-round, if you're a true die-hard)... well, it's worth the risks. Dealing with all the blue-hairs is a bit of a distraction, but you quickly learn not to stay behind cars that are larger than most boats.
And for people that will ignore an evacuation notice [...] Natural Selection
Actually, I think that most ppl who have lived here through a previous hurricane would agree with you. It's the non-natives who freak out the most. The beach-side hotels (on the barrier islands, at least) kicked everyone out Monday night, so even tourists have a fighting chance.
The radio this morning made a good point: the Sheriff's office is telling ppl to get out now because they won't be coming in to help them out later when it gets worse. If that's not a excellent case for your point then I don't know what is.
Man I hope KSC doesn't get damaged...all four Space Shuttles are there.
I note that the webcams at Universal Studios are toast ("temporarily unavailable due to preventative maintenance"). Oh well. I guess that means no footage of tourists fleeing for their lives from the big, bad storm.
(And to anyone who thinks I am being callous, bugger off. I live in Orlando with family in Melbourne, and am not worried because we are *prepared*. You live in Florida, you're gonna get stomped by a few hurricanes. Deal with it.)
Slashdot Mirror
I took the Core+OS version. From what I understand, they redid the structure after I took mine? Hopefully it's less of a joke now. Mine didn't have any technologies newer than USB 1.0 and Win2k.
Having just obtained both CCNA and A+ certifications within the last 9 months, even though neither was really necessary for my current job, I feel the need to defend the CCNA exam by contrasting it with the A+ exam.
The A+ exam is a complete joke. It might have been relevant 10 years ago, but isn't anymore. The vast majority of questions were completely irrelevant today, and mostly irrelevant for any computer faster than 500MHz. Most of it was pure memorization, the kind all those cram books are for. There were very few real-world questions -- less than a half-dozen. There were even a few questions that were nonsense -- none of the answers were applicable, much less valid, and in some cases the question didn't even make sense.
The CCNA, however, is a killer. I took mine in May 2007. The first part of the exam is roughly one-third memorization questions, one-third diagram interpretation questions, and one-third real-world questions. Most of the diagram questions are trick questions with multiple realistic-looking answers. (In other words, you can't just look for the "obvious" answer.) This part of the exam is meant to test your grasp of networking concepts.
The second part of the CCNA exam is what really gets you, though. It's all about configuration. Most of it is spent in a simulator. And not just a simulator for one router or switch, but a simulator for an entire network. One of my questions involved configuring 4 different routers and 3 switches. Oh, and they can disable parts of the simulator to make your job harder -- like having to diagnose a connectivity problem without being able to ping or traceroute. And yeah, they like to throw multiple IOS versions at you to make sure you know the different variations of the commands (especially for switches). You cannot cram for these simulator-type problems.
I tend to consider myself a pretty smart guy. I've been working as a network admin for 10+ years, albeit not with Cisco equipment. I aced the classwork for the CCNA courses without putting forth any effort whatsoever. I did homework in class and never had to come in after hours to catch up. And yet, I had ~45 seconds left on the timer when I finished part 2 of the CCNA exam. It's that tough, and they've got it timed down to the last minute. You do not have time to flounder and guess.
If you don't know your stuff backwards and forwards, you are not going to pass the second half of the CCNA exam. It's that simple.
Now, having said all of that, remember: the CCNA is the entry-level exam. It's not meant to certify that you can walk into a company and rewire an international infrastructure by hand. It's meant to certify that if you put me in front of a router or a switch or a small network that is having problems, I can most likely figure out what the problem is. The building-huge-networks stuff is part of the CCNP, not the CCNA. (The first CCNP class is, after all, "Building Scalable Networks".)
I see plenty of haterade about the CCNA exam, but I never seem to see it from people that have taken the tests. And I have to wonder: for all of those exam-crammers with CCNAs that everyone seems to know, when did they get their certs and are they current? I doubt it.
Disclaimer: I'm the guy that wrote TorrentSpy (the application, not the web site) and have contributed a small amount of code to the Python/core BT client and tracker. I haven't written any code for BT in a while, nor have I chatted with Bram in literally years, but ...
... but he also gets something he can brandish at anyone who wants to throw money at BT.
From what I remember, Bram always viewed the Python/core as a sort of "reference implementation" -- it was never his goal to make the Python client or tracker the end-all be-all.
Why is it then surprising that he'd want to bring on a client that doesn't have to be Open Source, and thus doesn't have to be clean and perfect, but is still sexy as hell? He still keeps his reference implementation that supports the features and is easy to reimplement a dozen times in two dozen languages
Remember that his goal since incorporating has been to legitimize and broaden the adoption of BT. A sexy client is a huge step towards that goal. It's not like Sony or the MPAA or whomever is going to distribute a customized version of Azureus any time soon -- it's a beast! But a custom version of uTorrent? A 1MB executable that you could throw on a CD that requires zero install? YTF not? Remember also that Strigeus has been working towards licensing out the core engine for uTorrent.
Plate. Shrimp. Plate of Shrimp.
Some of you are excessively paranoid. You know that, right?
(Yes, yes, I know: "Welcome to Slashdot".)
I have, right next to me on my desk, a stack of 1x CDRs that I burned back in 1996 on my double-height SCSI 1x CD burner. There are several different brands, but the majority of them are Maxell and Verbatim.
Each and every one of them still works perfectly. And no, I have not stored them vertically in a cool, dry place. Some of them were left on the front seat of my car, under the seats in my car, and various other random places.
On these CDs I have backups of web sites I used to work on, horrifically low-res quicktime video clips of Babylon 5, and a ton of MP3s. Yes, MP3s from 1996. No ID3 tags, and in ALL_UPPERCASE because of the Joliet/ISO FS wars from the time, but still perfectly listenable. Many of them ripped and converted on mindbogglingly slow 200MHz machines before MMX, and many of them downloaded over HTTP from web sites back before P2P was even thought of.
So, yeah, the article is bunk. My stack of nine-and-a-half year-old CDRs is working just fine, thank you.
This is a totally shameless self-promotion, but ...
I recently completed the first half of a book that features a program that does exactly this: plugs into your software and calcuates trust networks, then takes it one step further and applies firewall filters based on those networks. Basically, a P2P version of PeerGuardian, but smarter and more plugin-friendly. The author is Xochitl Green, a web programmer who quickly discovers that she's out of her league because there are people out there that don't necessarily want self-securing computers.
The first chapter of "Trust Network" is available for download from my site, and you can get the first half (up to chapter 18, about 150 pages) through lulu.com in either PDF or printed form, for what I think is pretty cheap. (There's even a large-print version for the visually-impaired.)
I'm pushing for the second half to be out by March 30. I don't claim to be Gibson, Gaiman, or Stephenson, but it's not Star Trek fanfic, either.
While I agree with the point you're trying to make, you're basing it on some assumptions that are not valid. First, dad isn't just a software geek, he was an ATC for 20+ years in the US Navy with a couple of years as an ATC in the private sector thrown in. He's used everything from flight strips and binoculars to the latest and greatest software. He's also been using computers at least that long and knows his way around every version of DOS and Windows and is even really getting into *nix. Hence why he is in the position he is in: he knows both sides of the game. But I know you weren't on the attack, and I'm not trying to be on defense, just clarifying a bit.
Secondly, the ATC folks definitely know what they are talking about. Hence why the software team is made up of a bunch of ATCs overseeing all the programmers and database weenies. This isn't some outsourcing gig we're talking about here. It's people redesigning their own world from the inside.
The point I was trying to make was not "throw out the old crap and make everything whiz-bang". Nor was I trying to paint the world over with Apple's HIG. I was attempting to point out that the system that is in place has become so esoteric and counterintuitive that it is actually hindering the training of this generation of ATCs. Or, like the topic says, "know your userbase".
When people have to make a conscious effort to rewrap their brain around some foreign thing every time they go to work or come home, it becomes stressful. Stress leads to mistakes. If you speak only Chinese at work but only use American Sign Language at home, don't you think you are going to slip up every now and then? Potential new ATCs have been raised on the same basic interface rules for the past 20+ years. Old ATCs have been trained on an orthogonal set of interfaces for at least as long.
The problem isn't that the interface that the old ATCs use isn't valid. Obviously it is valid, as it has worked for so long. The problem is that the interface they are used to (knobs, switches, dials, paper flight strips, etc) doesn't translate well to a computerized version, and even if you shoehorn it in, that computerized version doesn't feel right to younger kids, who are the whole reason for making the translation in the first place.
Remember how much controversy was stirred up when Apple made the first pretty version of Quicktime for Windows? It had jog dials and a draggable "drawer" that just didn't make sense to most people. Sure, those interfaces are fine for 3D-space, but for half of the people the digital versions suck. (Notice how both of those mis-features have been retooled since then?)
You mentioned that efficiency is the crucial element, training be damned. I agree. But I think you are missing the fact that the old ATCs are efficient not necessarily becuase the system is good or bad, but because they've been trained to be efficient at it. Potential new ATCs have been trained to be efficient at using current HIG. Why retrain them?
So what's the right answer? Build a system with a whiz-bang interface that any XP user off the streeet can figure out, but is nothing like what has been in place for the last century? Or build a system that people who have been ATCing for years can relate to, but is counterintuitive for everyone else and requires extensive mental conditioning to use? Or build a system that tries to mesh both paradigms but is still nonintuitive to both new and old ATCs? I certainly don't know.
My dad is helping the FAA and the US military design and roll out the next gen ATC software here in the US. He comes home and tells stories that make my skin crawl.
The first version of the software was built using standard current interface guidelines and widgets and the testing group that had no experience with older ATC systems were wowed at how simple and yet powerful it was. Pretty much any random person off the street could look at the screen and easily figure out what was going on and how to do various basic tasks. When that version was demoed to the ATC union the union freaked out at how different it was and thus began a cycle of making it more and more backwards.
So, nowadays the next gen ATC software almost exactly replicates the UI of the old non-computerized and semi-computerized systems. On-screen toggle switches and dials, that sort of thing. The FAA and the ATC union have decided that retraining all of their ATCs to use modern computer interfaces would be a Bad Thing. When the computer screen doesn't exactly replicate the interface of the 50+-year-old systems, they freak out and scream bloody murder. On the flip side, kids coming into the field today that have been using computers most of their lives are finding the interface to be counterintuitive to the point of being almost unusable. Middle-aged workers who are both highly proficient ATCs and home computer users report that switching between the two types of interfaces each night when they go home requires conscious effort on their part, since they are so orthogonal.
So who wins? Historical inertia, of course. Why fix the problem today when you can wait for your successors to fix it in 25 years?
Went to the web site and found out that my video card is specifically excepted from the compatibility list. I have a GeForce 4 MX 440 with 128MB of RAM. This card is what, 2 years old? It's certainly not a low-end card, to be sure. And, given the above metrics, the MX series is a good 25% of the user base. And yet, it doesn't have the all-wonderful pixel shaders and is somehow obsolete? I tried using 3D Analyzer to disable the pixel shader requirement for the game, but still no joy. But at least then I got an actual GPF instead of nothing!
But wait, here's the best part -- the SC:PT demo runs without a hitch on this card. That's a little strange, eh? Sure, I can understand that maybe the pixels shaders aren't needed until some end-level of the game that the demo doesn't have. But doesn't that imply that maybe the pixel shaders aren't really all that necessary to play the game? What ever happened to a nice graceful degradation of features? If I play Homeworld2 I can disable graphics features until I'm looking at fighting polygons. (Not that I have to, mind you, as HW2 worked out of the box on my system.)
Why not upgrade, you ask? Why should I? My card is still good enough for every other recent game I've tried to play. I'm not the type of gamer that needs to see individual strands of facial har whizzing by at 300 fps. I'm not saying we should go back to 2D sprites, but this is just ridiculous. If I were a tinfoil-hat-type, I'd be ranting about a nefarious plot by the gpu manufacturers.
So I gues my issues are thus:
You know, I've read the articles about how hard it is to get a decent game published. (Probably not in this case, given the publishers, but still.) I try to support the developers by buying a legit copy. I'm staring at a shelf full of game boxes. And yet, this kind of crap happens. While this isn't going to turn me off from buying games altogether, I doubt I'll be paying for any more Ubisoft games in the years to come.
SupaDupaSpy Syd and Noah Hicks (Peter Berg) are on a plane back from Madrid (?) and he asks her why she never met up with him in Rio (?). She's shocked and has no idea what he's talking about. He'd encoded the meeting invite in the headers of a forged spam email. She never got it because she has her computer automatically delete all spam.
/. is scooped by the pretend-CIA by over a year.
Pretty sad when
Just to be the voice of dissention ... I'm apparently the one person in the world for whom the PHP+Apache2 combo doesn't actually work right. Yes, it mostly works, but I'm constantly having to close out the windows that pop up when a thread goes boom. Yes, I am running it on XP (with all of the necessary service packs and hotfixes), but that shouldn't invalidate the fact that PHP+Apache2 isn't production-quality stable. And yes, I am 100% sure it is PHP causing the problem, as the errors only started when I started using PHP, and they increase as I convert more and more of my site to PHP.
Now, if I'm getting all these errors, why am I using it? Because it's still better than the alternative. When any DLL goes pop under IIS you get really flaky and esoteric things that start happening. Under Apache2 it just nukes a thread, which Apache2 diligently respawns and goes on with life. I can deal with clicking OK on a dozen windows a day if it means I don't ever have to worry about restarting my web service.
PHP+Apache2 isn't perfect, but it's good enough for what I need.
I reviewed this book on IBList.org last year. I figure I'll repost it here just to add another voice to the chorus.
-----
Economics, espionage, nanotechnology, black holes, social enginerring, and carnival phenomena. This book winds all these disparate subjects together into one (mostly) cohesive plot. This tends to lead to parts of the book reading more like college textbook excerpts than light sci-fi reading, but that may very well increase the appeal for the hard-core geek readers. The pacing occasionally suffers from the massive amounts of technobabbling exposition, but you still slog through it like a rubbernecker watching a car crash -- you just can't wait to see what it all means.
The character development is better than average, though there could have been more character-building scenes without significantly slowing the pace. Indeed, the technologies and concepts often get more ink than the characters do. (Because, really, there's only so often you can be hit over the head with the "socialism/marxism/communism/*-ism is bad!" bat before you're ready to start skimming instead of really reading.)
Overall, this was a good book. It could even make a good series, should Stross continue to write for it. College students pumped up on technobabble and economic/social theory will breeze through it, but the rest of us will still enjoy it.
The tracker already has all of this data. It has a count of how many people have completed as well as how many people are incomplete and how much they have left to go. (Like this, a CGI script that just reads the tracker's state file.) In fact, there's even a stats generator that makes pretty graphs out of it all. We're working on getting the UI elements to display these stats integrated into the core (example), but for now there are patches that do it. (Shameless plug.) To top it all off, the tracker generates log files that are plain old HTTP logs, and are parseable by pretty much any web stats generator.
If AOLTW/WB/etc ran their own tracker they'd have all the stats they needed. If you upload a torrent to a 3rd-party site and are the initial seed and want to see the stats, pretty much all of them nowadays let you at least see how many total downloads there have been. It's not an issue.
Another poster already explained some of the features you had questions about. As for the implied "why would I go through the extra step of having to configure the Launch button or use the Check button?" question ... the default BT3.x client does not as yet support the ability to check the current number of seeds and leeches (people with 100% of the file, and people with something less than 100% of the file), nor does it tell you which sub-files are complete/incomplete. Checking the seed/leech count is a quick HTTP operation so it is done by default when you open the app. Checking the files for completion, however, can take quite a long time for large files, and if it is a single-file torrent then you really don't get much more information than you would with the normal BT client. Therefore, it's a manual operation.
The way TSpy is intended to be used and is actually most commonly used is as a quick way to check the seed/leech count. This is because for most of the file trading that is currently done with BT, a torrent will stay active for a week at most and then go dead with no seeders. It sucks to launch the BT client, have it check your 2GB torrent, then wait an hour only to get stuck at 60%. TSpy can tell you in 5 seconds the odds of you getting a good download. The other usage that has become quite popular is for a seeder to pop up TSpy every now and then to see if anyone is still leeching. If not, they can take down their BT client and go seed something else. In both of these cases you don't necessarily want to launch the BT client, which is why that is also a manual operation.
The more granular file check is in place to help people recover something if it looks like their torrent is dead and they don't yet have all of it. It'll tell you which files are complete and how much you are still missing. And, before BT3.2 added in the download count counter, it was amusing to just keep rechecking the files to watch the bits pour in.
Admittedly, the extremely casual BT user probably won't have much use for TSpy. They'll probably be getting only the most popular and long-lived torrents that are almost guaranteed to always have seeders. But for the people that are trying to keep track of files that start and die out within a few days' time, TSpy can be rather useful.
HTH,
-KB
Yeah, it's buried way down in the body of the article, but it's still there. w00t! I'm the guy that wrote TorrentSpy. The web site that the link points to is running on my PC at home on cable 'net access. 300 hits in 3 hours. Not too bad so far, I guess. My gf is gonna kill me if I get slashdotted, tho. There are worse ways to go, I guess ...
I've had several people express interest in an *nix version of TSpy. It's written in Delphi, which theoretically means it shouldn't be too hard to port over to Kylix. If anyone has Kylix experience and would like to help out, feel free to contact me at the email on my web site.
I'm gonna go tail -f my log file and listen to my network bog down.
I think it is kind of funny for millions of people to live on a sand bar that gets hit over and over again by hurricanes
Yeah, but when you can walk out of your house in the morning and be standing on a sandy beach with the sun shining and warm water 9 months out of the year (or all-year-round, if you're a true die-hard) ... well, it's worth the risks. Dealing with all the blue-hairs is a bit of a distraction, but you quickly learn not to stay behind cars that are larger than most boats.
And for people that will ignore an evacuation notice [...] Natural Selection
Actually, I think that most ppl who have lived here through a previous hurricane would agree with you. It's the non-natives who freak out the most. The beach-side hotels (on the barrier islands, at least) kicked everyone out Monday night, so even tourists have a fighting chance.
The radio this morning made a good point: the Sheriff's office is telling ppl to get out now because they won't be coming in to help them out later when it gets worse. If that's not a excellent case for your point then I don't know what is.
Man I hope KSC doesn't get damaged...all four Space Shuttles are there.
They actually have a few webcams up: http://www.ksc.nasa.gov/ shuttle/countdown/video/video.html.
Why is the Space Center in Florida again?
Because even rocket scientists need a little sun.
I note that the webcams at Universal Studios are toast ("temporarily unavailable due to preventative maintenance"). Oh well. I guess that means no footage of tourists fleeing for their lives from the big, bad storm.
(And to anyone who thinks I am being callous, bugger off. I live in Orlando with family in Melbourne, and am not worried because we are *prepared*. You live in Florida, you're gonna get stomped by a few hurricanes. Deal with it.)