Augh! Don't give me this crap. I cannot believe that you're making some pointless distinction between "coder" and "programmer". Not to mention "engineer".
I am not. I am making a very important distinction between people who do simple programming that requires not many skills and other people who build complex software systems, which requires lots of skills. I just gave them different names ('coders', 'software engineers') for the sake of introducing some nomencalture, in accordance to the quote in question. My point was to explain what I think 'coders' in the original contect means and what not.
Coders, who implement fairly straighforward little programs according to existing specifications, are not necessarily incompetent, and it sometimes _is_ a good idea to hire a few coders that can so programmining on a rather low skill-level.
If we had less idiots saying "my role is an architect, not a coder" - or vice versa - then the software industry would be vastly improved.
Yeah, those are a pest. Those and the arrogant hackers who run around calling everyone they think they have to disagree with a 'popmpous prick'.
_Programmed_ is the keyword here. I'm sure you would have been able to get better at it if you had gotten some quality education and put enough time and effort in it; but you're right, writing good programs is a very complicated task. As your projects become bigger and more complex, you deal with software engineering, data engineering, all kinds of very academic stuff.
However, the quote mentions _coding_. Coding is not about writing high-quality software, it's about hacking together stuff like GUI frontends for simple database-driven business applications in a way that somewhat satisfies the customer and maximises the manufacturer's profit. Coders don't think about software architecture, that's what their bosses do. Coders are given specifications for small tasks and hack together some code that does approximately what the specs require, according to mostely rudimentary quality assurance testing. Coders generate heaps over heaps of cumbersome, hard-to-maintain, very redundant, error-prone code that could be easily replaced by a concise, reusable, highly configurable, transparent (as in easy to debug) implementation written by a good programmer.
However, it's mostly a non-trivial problem to find good programmers and pay them adequately, too. That's why most software is implemented (not necessarily planned) by bad coders who are indeed doing very low-skill work. And yes, that's one of the reasons why most software sucks.
The % of/. members who are in ACM must be very small because ACM only has 75,000 members in total.
My Uni is one (1) member of ACM, but all the students get access to the papers through it (i.e. when connected through the net on campus, a VPN or the Uni's own ISP). I guess there are a lot of organisations just like that.
Charging by the hour is tough when working alone because of the record keeping and the feeling that one will be accused of laziness for spending a lot of time at a *seemingly* simple task.
I reccommend negotiating a contract based strictly on an agreed upon task list (with a dollar amount affixed to each task).
Coming up with estimates of the difficulty of or needed time for a programming job can be difficult, because unlike in almost every other profession, you'll very likely be doing something that has never been done before. Therefore, it's common that people spend a lot of time at a task that *seems* simple initially. If you have an employer/customer that actually reads your records and discusses the project's progress with you regularly, they'll understand the situation and won't accuse you of lazyness. If they're not interested in what you're doing or too dumb to understand it, they can't find a proper reason for calling you lazy either.
Because of this difficulty, it's often very impractical to estimate a charge beforehand. You'll be afraid of demanding too much, because then they could give the contract to someone else who wants less, so you'll rather stick with your lower estimates and almost always end up working many more hours than estimated, for the same pay. Turns out charging by the hour isn't that bad an idea at all.
However, if this is a problem, there's a way around it: The pre-project. Offer to first do a quick&dirty prototype, a proof of concept, an analysis of the exact requirements, a UML diagram of the architecture; whatever helps you get to know the project in detail. Demand a flat fee and spend a fixed amount of time doing this. It doesn't have to be perfect; you can state, for example, "part foo of the data model is not finished because the responsible guy at the customer's company gave imprecise information". If you offer a detailed analysis of the project, based on which anyone, not just you, could implement the whole project, your customers will probably accept. After that, you have a much better idea of how much work it will be to finish the project and how long it will take you. And if you found out that the conditions are crap and you don't want to do the project, you can just walk away and have already earned money. It's still better to be payed by the hour IMHO.
Dear Natalie's Hot Grits, I'm afraid your attempt at being a smart-ass failed completely, because you simply got the facts totally wrong.
Your parent is talking about COPORATE OWNED MACHINES...
No, s/he is absolutely not. The comment referred to the situation of student-owned machines in a college dorm.
NOBODY is talking about doing this in EMAIL spread viruses, but with WORMS. You seem to have no idea the difference between the two.
I think it's rather you who doesn't know the difference between viruses and worms. Viruses infect other executables, while worms are self-contained. It's perfectly normal for worms to spread via e-mails and/or e-mail related infrastructure/protocols/software.
Anyway, my example about software that is distributed via e-mail and would have to be executed by a user him-/herself was about what I could (barely) accept. In the context of my comment, this was the exact opposite of the "WORMS" you believe to know so much about. I said...
I don't think it's right to write an anti-worm that gets into a system through a security vulnerability and runs without the user's consent and/or knowledge.
I think you missed the point: the anti-worm is deployed like a worm, but it is WELCOMED by the infectee. It is like inviting a little "angel" to come live on your system...
The problem is, worms don't wait for an invitation, per definition. They aren't software you want to execute -- if you wanted it, you'd download and execute it by yourself. You're totally right in saying that education is important. We should educate people that worms are dangerous and what they can do about them. And how often they affect non-Microsoft systems ^_~ But we shouldn't sneak unrequested software into their systems, which could potentially steal system resources during time-critical operations like burning a CD-R or submitting your homework at the last second. There's a reason why patch installers tell you to save all your work and close all other programs; omitting this is asking for trouble.
Why haven't antivirus companies caught on to this [anti-worms] ?
I think one of the reasons is that people don't always want to fight fire with fire. The Joe Average User types are now very slowly learning that malware such as worms is out there and that they should better protect their hardware. In this situation, telling them that there aren't only bad worms, but also good worms, wouldn't necessarily help.
For example, you can't teach them how to tell if an e-mail really came from Microsoft or if it's spoofed -- it's just too complicated. However, you can inform them of the fact that Microsoft never ever sends out patches in e-mails, so when they get an e-mail claiming to contain a security-relevant patch they should install, they know it can't be for real and is probably very dangerous.
If you distribute a hotfix like this, I might find that okay. After all, it'll only run on machines of users who were stupid enough to execute something that came in their e-mail. However, I don't think it's right to write an anti-worm that gets into a system through a security vulnerability and runs without the user's consent and/or knowledge. And, IANAL nor American, but I'm pretty sure it's against the law, too. That should be enough reasons why anti-virus companies haven't cought on to this.
Of course, the situation is different in a company that owns all the computers, as opposed to a college dorm. But OTOH, an admin that has access to all boxes on the net should be able to remotely make them install patches without resorting to worms.
With scripting kits, brains are not a requirement for writing a virus.
Your parent poster talked about a good virus. Of a good virus, I'd expect it to get past recently updated virus scanners at least in the first week of it's existance, instead of only compromising insecure boxes. You can't do that with a scripting kit (maybe unless you're the first user of this kit who releases his creation into the wild).
Dude... I've been running a well secured Linux firewall and a separate Linux based Apache web server since 2000. I've patched and updated the systems so far from their default install (RH 6.2) that they are no longer really RedHat boxes. These boxes have not been cracked once.
While I don't doubt that you set yourself up a very nice installation, this doesn't prove anything. For my parents' web surfing and my own gaming needs, I've been running two highly insecure Windows boxes (the second replacing the first when it grew too old) since 1993. I hardly ever patched and updated it (just new Windows versions and patches for _really_ frightening security holes). These boxes have not been cracked once either.
Well, light is continous, so you can get turquoise either by mixing B (450nm) and G (550nm), or by just have C (500nm) light.
(Without checking the wavelength numbers,) This is correct. However, I believe you got the reason for this phenomenon slightly wrong...
I guess that light is like a vector space: any base vectors will do as long as they are... crap, my memory sucks... independent?
Light itself is not like a vector space. If you produce turquoise light by mixing a blue and a green beam, you still have photons of two distinct wavelengths, which is fundamentally different from a truly turquoise beam with photons of one particular wavelength. This is why you can use a spectrogram to see, say, which elements a distant star consists of.
Light does, however, seem to be continuous to an eye or sensor, because a 450nm sensor doesn't just react to 450nm photons, but also to photons of similar wavelengths; just with decreasing sensitivity as the wavelength difference becomes greater. Turquoise light registers on both blue and green sensors, but hardly on red sensors, no matter how it is composed exactly.
Vectors in a colour model don't have to be orthogonal/independent (I think this term is correct, but I'm not a native English speaker). Especially sensors can be laid out in a redundant way; this is exactly the point with the new camera: sensors with overlapping areas of sensitivity, for more accuracy. However, it doesn't make much sense to store data in such a colour model, because the data would be redundant and thus waste more space than necessary and potentially contradict itself. But in different steps of visual reproduction (photography, image editing, printing) you mostly need to convert between different colour models, and by far not all of them are orthogonal. Maybe you know these 'photo' inkjets that have 6 different inks or even more.
The people who spell it "an historic" aren't pronouncing the "h". I say it and spell it the way you do, but AFAIK they're both valid pronounciations.
I don't think so. It might be okay to 'swallow' the H if it follows a consonant. For example, when a person talks rather quickly, not many would miss the H of 'historic' in this sentence: "It was the greatest historic event of that decade." However, you should hear the H when it follows a vowel, and 'a' is a vowel. It is only changed to 'an' if it precedes a vowel sound, which is not the case with 'historic'. And it also isn't, of cource, the case with 'university' ^_^
Slashdot really needs A "news that doesn't require discussion" section with comments disabled. Is there really anything worthwhile to say about this article?
So you were sure from the start that this article doesn't require discussion, because nothing worthwhile could be said about it. In spite of this, you wasted your time by first reading some superfluous comments and then complaining about them and demanding to gag users whose comments you don't want to read.
I wonder what the people who modded this "5, Insightful" were thinking??
You can't generalize about people. Though that's a generalization, isn't it?
Firstly, it's not a generalization. Look up the variousdefinitions of the word. I assume the one you mean is "reasoning from detailed facts to general principles"; however, the claim that comprises the first sentence doesn't contain any detailed facts to derive the stated general principle from. Secondly, there's nothing wrong with generalizing; the Bad Things are over-generalizations and faulty generalizations (I guess the first are a subset of the second), which you can't do if you're not generalizing in the first place.
one thing you have to keep in mind when you do this is that it's impossible to make a perfect plan beforehand -- it can change every time. if your plan says that task 2 will be done after task 1, and task one turns out to be delayed postponed due to previously unknown circumstances, you must absolutely avoid ignoring task 2 as long as number 1 is unfinished. this happened to me in the past, because i prioritized my to-do-list so the most important task would be done first, and i often made the mistake of thinking it wouldn't make much sense to do anythign else as long as the most important issue was still waiting to be dealt with. i have since learned that choosing what to do is actually quite difficult. always picking the top priority activity is wrong, because it could take hours or even days until you find the motication to actually finish it -- time in which you could accomplish other things. but it's also wrong to think any activity is okay, as long as it makes your to-do-list shorter, because a tidy, cleaned apartment and alphabetically ordered CD collection won't help you with the paper that's due tomorrow.
what works for me is talking to a person to which i'm important and who cares about me, and telling him/her of my most important issues in a way that i know they will ask me about them a few days later. this gives me motivation to get things done, so i can be proud when i tell him/her about them, and i'm also reminded to tackle tasks i haven't been motivated enough for yet.
Whether this is because you need a higher income to own a Mac or because using a Mac leads to a higher income, is undetermined.
This implies that it would have to be one of these options, which is wrong. Don't mix up coincidence with causality. Anyway, how did you arrive at the conclusion that
"
Mac users are more likely to be republicans"
in the first place? Any facts to back that up, or just trolling?
CDs have almost perfect channel separation, so you could put one side of the tape on the left channel, and the other one on the right.
Isn't the channel separation of a tape also good enough for the low data rate of C64 programs? Yes, doubling the capacity of a cassette this way would be a custom hack -- just like your CD idea.
Among other places where this scheme is legally questionable, the rules explicitly prevent radio stations from doing things like allowing listeners to democratically select which songs to play.
Correct. They are also not allowed to publish playlists of future broadcasts; at least in my country, but I guess that's about the same in the US. What you could do without violating the law would be setting up software that constantly monitors the webcast of a radio station and rips all songs that match a keyword from a list you defined, or similar criteria. You could be content with having found a station you like, which broadcasts lots of music you want to keep, or go further and send them some requests, hoping that these songs will end up on your hard disk sooner or later. You don't even need to get the stream from the web, you could get it off the air and have some custom software reference the songs with the station's on-line playlist.
But what you absolutely cannot do is legally disguising peer-to-peer filesharing as radio broadcasts.
Thanks! I was so sure there wouldn't be a FreeBSD driver that I didn't even look, shame on me ^_^. Lately, I never knew how to choose between nVidia and ATI -- I'd say this problem is solved now.
Nvidia supplies us binary drivers, but at LEAST they supply us drivers!
This is not a rhetorical question, I'd really like to know: What if I'm using FreeBSD (to which I'm considering to switch at home because that's what my web server runs), would a binary Linux driver help in any way or be useless?AFAIK, there's something that let's you run Linux programs on FreeBSD (without recompiling), but I guess that doesn't apply to drivers?
With emails, people with a lot of time on their hands can, obviously, give themselves room to develop an ironic theme, but for people with jobs, e-etiquette demands instant response, which brings you down to the very rudiments of irony - I Love My Boss; I'm Delighted That My Ex Is Going Out With That Attractive Woman; I Really Couldn't Be More Pleased That You've Lost a Stone.
I don't want to object that these aren't fine examples of rudimentary irony, but one could argue that they are mainly sarcastic. Zoe Williams laments that irony is often mixed up with hypocrisy, cynicism, laziness, and coincidence, but completely fails to mention sarcasm. Maybe this isn't a severe omission in the context of this article, because many more sarcastic statements actually show features of irony as there are ironic statements you could consider sarcastic.
Haven't read the article yet, but I also thought of Stephenson's 'Snow Crash' when I read the posting. For those who haven't read the book, it describes an ancient culture that used clay tablets to write down algorithms that would be executed by humans. Much like cake recipies, but the ruler/priest would decide what needed to be done (harvest wheat, build a house, depending on the season) and make the subjects 'run' the right 'script'.
And for those who'd like to understand the joke above...
--- SPOILER ALERT! SPOILER ALERT! ---
Snow Crash, in the book of that name, is a virus that infects programmers if they just look at a certain document.
They have renamed USB 1.1 to USB Full-Speed and USB 2.0 to USB Hi-Speed and use those new names consistantly throughout their web pages.
Web pages are changed quickly to be all nice and tidy. But no matter what usb.org says, if you google around a little, you'll find tons of products that are labeled USB 2.0, yet not capable of transfer speeds greatet than 11 MB/s. However, to this day, I would have expected all products labeled USB 2.0 would support much faster speeds; thus, I found this article very informative and interesting. I don't think it contains wrong/untruthful information and absolutely don't see indications that 'the author must have been smoking something'.
Oh, and if you look around a little on the consistant usb.org site, you'll find that several press releases, for example, contain terms like 'Hi-Speed USB 2.0' (not just Hi-Speed USB).
Slashdot Mirror
Coders, who implement fairly straighforward little programs according to existing specifications, are not necessarily incompetent, and it sometimes _is_ a good idea to hire a few coders that can so programmining on a rather low skill-level. Yeah, those are a pest. Those and the arrogant hackers who run around calling everyone they think they have to disagree with a 'popmpous prick'.
However, the quote mentions _coding_. Coding is not about writing high-quality software, it's about hacking together stuff like GUI frontends for simple database-driven business applications in a way that somewhat satisfies the customer and maximises the manufacturer's profit. Coders don't think about software architecture, that's what their bosses do. Coders are given specifications for small tasks and hack together some code that does approximately what the specs require, according to mostely rudimentary quality assurance testing. Coders generate heaps over heaps of cumbersome, hard-to-maintain, very redundant, error-prone code that could be easily replaced by a concise, reusable, highly configurable, transparent (as in easy to debug) implementation written by a good programmer.
However, it's mostly a non-trivial problem to find good programmers and pay them adequately, too. That's why most software is implemented (not necessarily planned) by bad coders who are indeed doing very low-skill work. And yes, that's one of the reasons why most software sucks.
Having said that, I'll go and RTFA now
the chain is called 'Lucky Dragon' and appears in gibson's 'all tomorrow's parties'
Because of this difficulty, it's often very impractical to estimate a charge beforehand. You'll be afraid of demanding too much, because then they could give the contract to someone else who wants less, so you'll rather stick with your lower estimates and almost always end up working many more hours than estimated, for the same pay. Turns out charging by the hour isn't that bad an idea at all.
However, if this is a problem, there's a way around it: The pre-project. Offer to first do a quick&dirty prototype, a proof of concept, an analysis of the exact requirements, a UML diagram of the architecture; whatever helps you get to know the project in detail. Demand a flat fee and spend a fixed amount of time doing this. It doesn't have to be perfect; you can state, for example, "part foo of the data model is not finished because the responsible guy at the customer's company gave imprecise information". If you offer a detailed analysis of the project, based on which anyone, not just you, could implement the whole project, your customers will probably accept. After that, you have a much better idea of how much work it will be to finish the project and how long it will take you. And if you found out that the conditions are crap and you don't want to do the project, you can just walk away and have already earned money. It's still better to be payed by the hour IMHO.
Anyway, my example about software that is distributed via e-mail and would have to be executed by a user him-/herself was about what I could (barely) accept. In the context of my comment, this was the exact opposite of the "WORMS" you believe to know so much about. I said
For example, you can't teach them how to tell if an e-mail really came from Microsoft or if it's spoofed -- it's just too complicated. However, you can inform them of the fact that Microsoft never ever sends out patches in e-mails, so when they get an e-mail claiming to contain a security-relevant patch they should install, they know it can't be for real and is probably very dangerous.
If you distribute a hotfix like this, I might find that okay. After all, it'll only run on machines of users who were stupid enough to execute something that came in their e-mail. However, I don't think it's right to write an anti-worm that gets into a system through a security vulnerability and runs without the user's consent and/or knowledge. And, IANAL nor American, but I'm pretty sure it's against the law, too. That should be enough reasons why anti-virus companies haven't cought on to this.
Of course, the situation is different in a company that owns all the computers, as opposed to a college dorm. But OTOH, an admin that has access to all boxes on the net should be able to remotely make them install patches without resorting to worms.
Light does, however, seem to be continuous to an eye or sensor, because a 450nm sensor doesn't just react to 450nm photons, but also to photons of similar wavelengths; just with decreasing sensitivity as the wavelength difference becomes greater. Turquoise light registers on both blue and green sensors, but hardly on red sensors, no matter how it is composed exactly.
Vectors in a colour model don't have to be orthogonal/independent (I think this term is correct, but I'm not a native English speaker). Especially sensors can be laid out in a redundant way; this is exactly the point with the new camera: sensors with overlapping areas of sensitivity, for more accuracy. However, it doesn't make much sense to store data in such a colour model, because the data would be redundant and thus waste more space than necessary and potentially contradict itself. But in different steps of visual reproduction (photography, image editing, printing) you mostly need to convert between different colour models, and by far not all of them are orthogonal. Maybe you know these 'photo' inkjets that have 6 different inks or even more.
I wonder what the people who modded this "5, Insightful" were thinking??
Oops, now that was off topic
one thing you have to keep in mind when you do this is that it's impossible to make a perfect plan beforehand -- it can change every time. if your plan says that task 2 will be done after task 1, and task one turns out to be delayed postponed due to previously unknown circumstances, you must absolutely avoid ignoring task 2 as long as number 1 is unfinished. this happened to me in the past, because i prioritized my to-do-list so the most important task would be done first, and i often made the mistake of thinking it wouldn't make much sense to do anythign else as long as the most important issue was still waiting to be dealt with. i have since learned that choosing what to do is actually quite difficult. always picking the top priority activity is wrong, because it could take hours or even days until you find the motication to actually finish it -- time in which you could accomplish other things. but it's also wrong to think any activity is okay, as long as it makes your to-do-list shorter, because a tidy, cleaned apartment and alphabetically ordered CD collection won't help you with the paper that's due tomorrow.
what works for me is talking to a person to which i'm important and who cares about me, and telling him/her of my most important issues in a way that i know they will ask me about them a few days later. this gives me motivation to get things done, so i can be proud when i tell him/her about them, and i'm also reminded to tackle tasks i haven't been motivated enough for yet.
But what you absolutely cannot do is legally disguising peer-to-peer filesharing as radio broadcasts.
Thanks! I was so sure there wouldn't be a FreeBSD driver that I didn't even look, shame on me ^_^. Lately, I never knew how to choose between nVidia and ATI -- I'd say this problem is solved now.
And for those who'd like to understand the joke above
--- SPOILER ALERT! SPOILER ALERT! ---
Snow Crash, in the book of that name, is a virus that infects programmers if they just look at a certain document.
Oh, and if you look around a little on the consistant usb.org site, you'll find that several press releases, for example, contain terms like 'Hi-Speed USB 2.0' (not just Hi-Speed USB).