i think its a solid language.
but its Microsoft's Language.
they have never done anything to earn trust.
atleast sun has.
Some day i will try to comprehend how anyone can say this. Do you guys not realize that Sun has NOT allowed Java to be submitted to open standards bodies, and instead chooses to keep it proprietary - while microsoft has already submitted C# and fully published it as an open spec?? Since this is the most relevant example of "earning trust" I can think of for the sake of this discussion, I can't imagine what more you need.
there are valid criticisms, but you haven't found them yet.
very well put - speaking as someone with a decent amount of.net experience, i couldn't agree more with this statement. the grandparent poster has clearly not put in the time to understand the true problems with c# that are encountered every day in real-world situations. i'm not sure if the parent poster has encountered these, but he clearly recognizes that they exist, and that the grandparent poster is focusing on hypothetical situations that are not typical of the real-world every-day situations that are encountered in.net development.
grandparent - if you'd like to build up your library of flame/trolling material, have a look at the following:
* typed exception handling (i.e. "throws" keyword - or lack of it)
* boxing (and the faults it introduces)
* remoting
* packages vs. assemblies, internal vs. friend
* etc...
however, before you get too hyped up on your newly constructed library of flames, you should also look into some of the improvements coming in future versions of.net, such as indigo, avalon, etc., to make sure your complaints are not already being addressed.
If i start a business importing these storage devices from France and Canada to the US, do my customers in America then get to share the same copy liberties as defined in these foreign laws, assuming they have paid the appropriate taxes?
What is the tax rate on cd's, hard drives, etc.? I'm sure there is a customer base that would be interested in paying an additional 10%-20% to have these liberties...
Don't pretty much all of us who graduated in CS and took CS372 learn operating systems by studying some real operating system code and concepts? And isn't it likely that most of us studied a unix or unix-variant codebase? So are you saying that only Liberal Arts majors are hired to work on Windows?
So for some reason in America Group A can get laws passed to ban the research. However isnt religious oppression illegal in the US? So why dont the lawyers that represent those companies fight it on grounds of religious oppression?
You've got the right term, but the wrong interpretation (based on the constitution and other precedents/laws). "Religious opression" is meant in this context to say that it is illegal for the U.S. government to "oppress" the religious beliefs of it's citizens - not the other way around, and not between citizens, other than that a citizen cannot oppress another citizen because of their religious beliefs.
This isn't France, you know. Seriously - I was just having this discussion the other day with a couple of French colleagues - apparently, they just passed a law in France that says that students (who are citizens - not government employees) in public schools cannot wear head wraps indicating their status as conservative Muslims. When talking to my colleagues, my understanding is that the intention here is essentailly what you are saying - that the citizens feel oppressed by other citizens imposing their religious beliefs. The constitution in France does not have the same phrasing as the one in the U.S. on this issue, and since the issue is left vague enough, laws like this can be passed in France. Ironically, it is likely that the original intention of these clauses in the French constistitution was originally the same as the intention of America's "founding fathers" - however, because of slightly different wording, these clauses have now come to have almost exactly opposite interpretations. Which is correct? Well, the one that's consistent with the constitution of your respective nation, I guess. Thankfully, in the U.S., our constitution is clear enough on this issue that it will never be interpreted in a way that allows the rule to be turned around against those with religious beliefs, since that was, after all, one of the founding principles of this nation.
Ironically, in France it is not illegal to wear a cross or other religious symbol in school, even if the cross is outside the shirt and relatively ostentatious. Clearly, there are some inconsistencies in the interpretations over there...
Original trilogy, SE on DivX. Sound is prologic, one disk per movie.
I believe there are also some 2-disc DivX copies of each of the originals (non-SE) out there somewhere, widescreen, and quality is rumored to be excellent, presumably ripped from LD or something.
This must be the ultimate Linux "use whichever side supports your current argument" statement. I can't guess how many times I've heard an argument like "actually, if you want to talk about older, more experienced OS's, Linux (Unix) has been around for 50 years" or whatever...
So now, just so we're clear, it is valid to make this argument when talking about an OS, but not when we are talking about the window manager?
You know, contrary to popular belief here, the actual consumer public doesn't care about the difference between the two.
I tend to agree with those that say if they can't make it work, they just aren't doing it right - companies are using the internet to ensure extremely high-profile transactions, and yes there are differences in verification, etc. - but the basic security principles are the same.
However, I also understand why the idea shouldn't be pushed out before it's ready. What I don't get is - why not an "official-unofficial" beta pilot program, where everything is provided as it might be the next time around in a production scenario, and internet votes will be tallied, but in the end those votes will not actually be counted towards this year's election - at least this would be a way to introduce the technology and the new system to the public, and to monitor for any problems. This would also potentially be a way to guage to some extent the level of security and the demand for attacks. But I suppose if the hackers knew it wouldn't count, they would have no incentive to attack...
Still, it seems a shame to simply say "we are throwing away the idea".
wow, i'm disappointed that i just used up my mod points for the day. this would have definitely been +1 insightful.
i really don't understand why everyone has to hate the new thing, like enterprise. i'm sure it was the same with TNG during the first couple of seasons, but probably even more so - for some people i think it still is!
maybe i just get wrapped up in trek stuff too quickly, and am too easy to please - but i for one consider it a sort of legacy, and when the legacy is continued and produced as well as it has been, well - that's something that makes me thirst for more.
i can't get over thinking about an episode i just watched the other day, i think maybe Enterprise 3x01 - there was a sort of enemy round-table scene, very similar (kind of a rip-off probably) to the separatists roundtable scene in ATOC, with CG aliens speaking strange languages and everything - and i was shocked because it was a tv show, with a scene at almost (about 90%) the same quality level as the equivalent scene in ATOC, which i consider to be one of the best effects movies of all time. i think shows like enterprise are not only carrying on the legacy, but also in some ways pushing the tv domain further in terms of production quality.
IANAL, but my understanding is that in the U.S., and I would assume other places such as U.K. as well, theft is a criminal matter, while copyright infringement is a civil matter - two completely different areas of the law, regardless of the "similarities". I believe this is part of the reason that slashdotters often disagree with the usage of the word "theft" in conjunction with copyright law.
If the barrio (filled with illegal border-crossers) just a few blocks away from my parent's neighborhood in HOUSTON, TEXAS is still part of Houston, TX, then why aren't the residents there citizens of the United States?
clearly you have not taken the tube into the city in london at 8am on a weekday - that is most likely the type of scenario the original poster is referring to. yes, almost everyone in the train has a laptop, but no, there is virtually no room to budge, much less to pull out a 14" and prop it on your shoulder. even if you are lucky enough to get a seat, you won't be in the mood to pull out a laptop and get it powered up. a portable video device (with headphones) would be optimal for this scenario, and i'm sure a significant percentage of the thousands who do this in london every day would be interested.
Sorry, not meaning to take it out on you specifically - far too many people are quickly throwing an analogy out like this without thinking, and you just happened to be the convenient one at the top to reply to. ;-)
A lot of people here are making analogies about the former, while this discussion is supposed to be focused on the latter. People are confusing the two as if they were the same thing. I entirely agree that MS products have far too many bugs, but that is not relevant to whether it is insecure. Granted, the two are related, but an analogy about defects is not implicitly applicable to security, and people are using this car analogy as if it were.
Thank you, this was exactly my point. I recall watching a saturday morning special or something about this - that the architects had tried to anticipate such a thing, but not fully accomplished the level of "security" that they intended, because there were secondary issues that they had not anticipated - much like our "insecure" software. Thanks to all who got my point - the building thing may not be the perfect analogy in this case, but it is certainly better than a "car".
Is it the driver's fault for using a car that explodes when rear-ended?
It never ceases to amaze me that people continue to make this ridiculous car/software argument. the correct analogy is:
Is it the driver's fault for using a car that explodes when a rocket-propelled-grenade is launched at it?
Rear ending a car is a normal day-to-day *ACCIDENT* that normal drivers do all the time.
Writing a virus/worm/exploit to specifically attack known holes in a piece of software is not an accident, and it is not something normal users do. Show me a RPG-proof car, and then I'll build you some hacker-proof software.
Face it people: the truth is that today, software products are far more "secure" than the majority of consumer products. You are just having a hard time making the distinction between warranty failures and security attacks. We are talking about security here, not faults. I bet a lot more cars get stolen every day than OS's and email clients that get hacked.
I don't mean to be politcally distasteful here, but instead of comparing a piece of software to a car, perhaps we should be comparing it to something like a building. Both have architects, so this makes sense. An important, highly-populated building should probably be designed to anticipate and withstand a variety of terrorist attacks, such as planes flying into it. Recent events show us that this is not always the case. And that cost us lives, not just dollars! But we don't constantly sit here and hate the architects - we learn to deal with it and try to find ways to get better, knowing that we will never completely eliminate all vulnerabilities the building might have to a terrorist attack. Today I don't think anyone walks into a building assuming the building is 100% invulnerable. Or a plane. Or any other highly complex, highly targeted construction.
Making analogies to cars worked when we were complaining about defaults. But it doesn't make any sense when we are talking about security. Let's try to use some better analogies, and think about whether they are relevant.
thanks for the user page tip - i now realize that my profile is already set up for this, but unfortunately i've been receiving all my notifications on my hot[spam]mail account... i'll have to adjust that...
i'm finding that at least philosophically, we have more agreement than we have disagreement - i certainly don't debate your feelings about what is "interesting" in the software world, or who is able to "accurately" predict software value - for the most part, i entirely agree.
my point about valuations is not to imply that i believe the method to arrive at those valuations is anywhere near correct or accurate - but the fact remains that while a language like C# may be nothing really new under the sun, there is still "value" in learning it for some (a lot of) developers, because those developers realize that that skillset == $$$ in their particular cases. in this case, developer income == the value or "worth" in learning the new language.
As a result, I don't see that.NET is doing anything new.
While the list you've given contains a number of wonderful names, I still stand by my point. The point is that 95% or more of the world's paid developers today are split between Java, VB, and C++ - and maybe Cobol and Fortran still make up a big chunk, but even then the point still stands...
Thus the "interesting" thing in this case is not a "software-interesting" thing - it is a "marketshare-interesting" thing - MS has released into the programming world the first set of tools *feasably capable* of meeting the needs of a vast majority of developers out there today.
To do this, you can't just have Java, a bunch of legacy languages, and a bunch of revolutionary new paradigm and scripting languages - you must also have the other staples of the market - thus you must have VB and C++. If for no other reason, this is because stubborn C++ guys are never going to give up, and a miserable majority of VB developers are sadly never going to make it to another environment before they have a career change.
Not the "interesting" you were hoping for? Hmm, maybe not - maybe not for me either - but for my team which may consist of a few hopeless VB and C++ guys, it makes all the difference.
I think you're on to the most important issue here, and everyone seems to be overlooking it. The biggest impact a ban like this can have is to give spammers insurance that they won't have to worry about this sort of anti-spam tax in the future. You are definitely right that that is (was?) one of the most promising schemes anyone has thought of to stop spam.
*sigh* oh well, at least I won't have to pay taxes... I can feel so happy to know that the $4.27 a month I'm saving by not paying these taxes only costs me 10 hours a month of sifting through spam!
rjh, Not sure if you're still following this thread - if you have some kind of auto-email thing turned on let me know, because I'm a relative/. newbie and have no idea how to set that up if it exists...
Thanks for taking the time to continue this discussion and write such a thought-out post.
It's meant to say that in hindsight many things become obvious which required real genius to see them in the first place.
Totally agree, and in the case of graph theory, as I said, Dijkstra definitely seems to have made a significant contribution, even if many of his findings do now seem somewhat obvious. What I really wanted to point out was the fallacy of the "language worth learning" quote, in that it only uses "worth" in once sense of value, when in fact the majority of people in the world who deal with technology use a completely different set of criteria on which to place the "value" of learning a language - and this usually comes down to $ in the end.
My insistence on multiple paradigms is unequivocally not born out of "academia", a word which it seems you're using as an insult. My insistence on multiple paradigms is born out of my demand that I use the right tool for the job.
Let me apologize as I see that it probably sounds like that's the connotation I'm putting on this word. I absolutely don't mean that - I simply mean that in this particular discussion, I am talking more about industry values, because that's what seems most relevant. I am all for the academic interest in alternative paradigms, and to a certain extent I fully agree with what you're saying here - that paradigms are a part of the "right tool for the job" decision. I simply don't believe I've seen this to be the case as much in real-world scenarios. I would love to have a particular area of the projects I work on which is best suited to Lisp, and at the same time to have the appropriate toolset to do so - but today it just isn't happening. Granted - one of the previous companies I worked with was a civil engineering firm, and the architecture of their systems was all about putting the engineering/mathematical models in Fortran, and then integrating that with object-oriented and COM tools on the front end and business layers - but I honestly believe the choice for Fortran was primarily a result of (1) applicable skillsets and low budget for training/upgrading of skillsets; and (2) technical capabilities of the particular toolset (Digital Fortran) in relation to certain mathematical optimizations (i.e. optimized array/matrix calculations); certainly these are great reasons in the real-world, but in what I'm calling the "academic" sense, there is actually nothing really interesting or even nice about this particular cross-paradigm use - if they had the same skillset level and also the same level of mathematical optimizations in a Java toolset, my academic sense tells me that the applications we built would have been much more nicely architected because there is no reason those parts of the application couldn't have taken more advantage of an object-oriented environment.
Interestingly enough, on the project I am currently involved in, a major strategic part of the project is a business-level "intelligence" layer, which essentially should allow for the development of intelligent strategies to manipulate interaction with the market. Our team made an attempt to build this layer to enable the strategies to be developed fully in an alternative paradigm language. The language is called ILog, and from what I understand, it is some sort of rules-based language, essentially based on pre-conditions and post-conditions, causing certain rules to fire. Unfortunately I wasn't involved on that part of the project, so I don't know much more about it.
But suffice it to say that this is a case where I can certainly validate your claims - extreme challenges were found when trying to integrate an object-oriented.NET architecture with an engine based on a different par
First, I'll ask you to please see my comments in the other response to you...
As Dijkstra famously said, "Any language which does not change the way you think about programming is not worth learning."
Ah, classic Dijkstra. At the risk of becoming flamebait for millions of CS-Drones, let me tell you a little something about him - I had him as a professor one semester - he was not always in touch with the real world, aspecially as he got older and the world got older - apparently they seemed to take divergent paths. He may have been dead on in the 70s when he "ingeniously" realized that GOTO was a bad idea, and maybe there was a better way. In fact, even his "brilliant" graph algorithms may have been top-class during his time.
But the statement above is a classic Djikstra statement, and although I don't specifically recall it, I have no doubt he said it, because it embodies exactly who he was. If in your academic mind you truly believe this, then that's good for you - you have an interesting career ahead of you, always looking for the next paradigm. But in the real world, there are many languages in each paradigm, and sometimes the reason it is "worth" learning a new language is not for academic interest - sometimes it is productivity of toolsets supporting that language, or ability of integration with other technologies, or *shock!* money! In the real world, there are many reasons it might be "worth" learning a new language, even if it's just like another one you already know.
As it relates to.NET, when people say it supports multiple languages, what they are talking about it the real-world effect of that - the idea is that if I have 2 teams with different skillsets (i.e. one with VB background, one with Java background, but both with new "upgraded".NET skillsets), I can still have them working together in harmony, with little integration overhead, because.NET provides *this* type of cross-language integration. This whole cross-paradigm thing is extremely interesting, but given that the major market of developers.NET has to target right now are split between C++, VB, and C# - since all of those can be adapted to be OO languages, the market does not really dictate that MS must focus right now on building cross-paradigm tools.
Re:quote: "the CLR integrates all prog. languages"
on
Does C# Measure Up?
·
· Score: 1
CLR and.NET are not multilanguage in any useful or interesting sense. In order to be multilanguage in a useful and/or interesting sense, it would have to support a wide variety of programming paradigms.
This statement may be true in your academic mind, but in the real world,.NET is clearly multi-language for all practical purposes. Just because you refuse to make a distinction between "cross-language" and "cross-paradigm", it does not follow that no such distinction exists..NET supports more than one language, hence it is multi-language. How much simpler can it be?
In regards to the rest of your comments regarding more historically interesting languages, and their likelihood of sticking around for a while - I am all on board with you. Lisp has long been one of my favorite languages, primarily because for me (and I'm sure others as well) a way of recursive thinking suits me better. I certainly won't disagree with you on the academic points, nor on your points about what may remain in the distant future. However, in the real world, and in the domain where this discussion started, the here and now is what was supposed to count.
The day I can find a job working on Lisp projects as interesting (in a business sense, not an academic sense) as the projects I work on in.NET are, let's continue this discussion.
Re:quote: "the CLR integrates all prog. languages"
on
Does C# Measure Up?
·
· Score: 1
Yes, but in the same phrase it then says "...allows objects..."
Clearly the domain of "languages which can understand objects" is implied. furthermore, although this is not written in the most straightforward way, "all languages" is clearly meant to imply all languages in the domain of ".NET-enabled programming languages".
In any case, I don't care much for this marketing spheel. I was under the impression that we were discussing the actual real-world multi-language capabilities of the CLR. Clearly.NET does not support "all programming languages", so any argument derived from a marketing statement based on that premise is flawed from the start.
Slashdot Mirror
I'm glad to see someone posted that - that's exactly how I read it.
i think its a solid language.
but its Microsoft's Language.
they have never done anything to earn trust.
atleast sun has.
Some day i will try to comprehend how anyone can say this. Do you guys not realize that Sun has NOT allowed Java to be submitted to open standards bodies, and instead chooses to keep it proprietary - while microsoft has already submitted C# and fully published it as an open spec?? Since this is the most relevant example of "earning trust" I can think of for the sake of this discussion, I can't imagine what more you need.
there are valid criticisms, but you haven't found them yet.
.net experience, i couldn't agree more with this statement. the grandparent poster has clearly not put in the time to understand the true problems with c# that are encountered every day in real-world situations. i'm not sure if the parent poster has encountered these, but he clearly recognizes that they exist, and that the grandparent poster is focusing on hypothetical situations that are not typical of the real-world every-day situations that are encountered in .net development.
.net, such as indigo, avalon, etc., to make sure your complaints are not already being addressed.
very well put - speaking as someone with a decent amount of
grandparent - if you'd like to build up your library of flame/trolling material, have a look at the following:
* typed exception handling (i.e. "throws" keyword - or lack of it)
* boxing (and the faults it introduces)
* remoting
* packages vs. assemblies, internal vs. friend
* etc...
however, before you get too hyped up on your newly constructed library of flames, you should also look into some of the improvements coming in future versions of
If i start a business importing these storage devices from France and Canada to the US, do my customers in America then get to share the same copy liberties as defined in these foreign laws, assuming they have paid the appropriate taxes?
What is the tax rate on cd's, hard drives, etc.? I'm sure there is a customer base that would be interested in paying an additional 10%-20% to have these liberties...
Don't pretty much all of us who graduated in CS and took CS372 learn operating systems by studying some real operating system code and concepts? And isn't it likely that most of us studied a unix or unix-variant codebase? So are you saying that only Liberal Arts majors are hired to work on Windows?
Hmm...actually, you may have a point...
So for some reason in America Group A can get laws passed to ban the research. However isnt religious oppression illegal in the US? So why dont the lawyers that represent those companies fight it on grounds of religious oppression?
You've got the right term, but the wrong interpretation (based on the constitution and other precedents/laws). "Religious opression" is meant in this context to say that it is illegal for the U.S. government to "oppress" the religious beliefs of it's citizens - not the other way around, and not between citizens, other than that a citizen cannot oppress another citizen because of their religious beliefs.
This isn't France, you know. Seriously - I was just having this discussion the other day with a couple of French colleagues - apparently, they just passed a law in France that says that students (who are citizens - not government employees) in public schools cannot wear head wraps indicating their status as conservative Muslims. When talking to my colleagues, my understanding is that the intention here is essentailly what you are saying - that the citizens feel oppressed by other citizens imposing their religious beliefs. The constitution in France does not have the same phrasing as the one in the U.S. on this issue, and since the issue is left vague enough, laws like this can be passed in France. Ironically, it is likely that the original intention of these clauses in the French constistitution was originally the same as the intention of America's "founding fathers" - however, because of slightly different wording, these clauses have now come to have almost exactly opposite interpretations. Which is correct? Well, the one that's consistent with the constitution of your respective nation, I guess. Thankfully, in the U.S., our constitution is clear enough on this issue that it will never be interpreted in a way that allows the rule to be turned around against those with religious beliefs, since that was, after all, one of the founding principles of this nation.
Ironically, in France it is not illegal to wear a cross or other religious symbol in school, even if the cross is outside the shirt and relatively ostentatious. Clearly, there are some inconsistencies in the interpretations over there...
Original trilogy, SE on DivX. Sound is prologic, one disk per movie.
I believe there are also some 2-disc DivX copies of each of the originals (non-SE) out there somewhere, widescreen, and quality is rumored to be excellent, presumably ripped from LD or something.
This must be the ultimate Linux "use whichever side supports your current argument" statement. I can't guess how many times I've heard an argument like "actually, if you want to talk about older, more experienced OS's, Linux (Unix) has been around for 50 years" or whatever...
So now, just so we're clear, it is valid to make this argument when talking about an OS, but not when we are talking about the window manager?
You know, contrary to popular belief here, the actual consumer public doesn't care about the difference between the two.
I tend to agree with those that say if they can't make it work, they just aren't doing it right - companies are using the internet to ensure extremely high-profile transactions, and yes there are differences in verification, etc. - but the basic security principles are the same.
However, I also understand why the idea shouldn't be pushed out before it's ready. What I don't get is - why not an "official-unofficial" beta pilot program, where everything is provided as it might be the next time around in a production scenario, and internet votes will be tallied, but in the end those votes will not actually be counted towards this year's election - at least this would be a way to introduce the technology and the new system to the public, and to monitor for any problems. This would also potentially be a way to guage to some extent the level of security and the demand for attacks. But I suppose if the hackers knew it wouldn't count, they would have no incentive to attack...
Still, it seems a shame to simply say "we are throwing away the idea".
wow, i'm disappointed that i just used up my mod points for the day. this would have definitely been +1 insightful.
i really don't understand why everyone has to hate the new thing, like enterprise. i'm sure it was the same with TNG during the first couple of seasons, but probably even more so - for some people i think it still is!
maybe i just get wrapped up in trek stuff too quickly, and am too easy to please - but i for one consider it a sort of legacy, and when the legacy is continued and produced as well as it has been, well - that's something that makes me thirst for more.
i can't get over thinking about an episode i just watched the other day, i think maybe Enterprise 3x01 - there was a sort of enemy round-table scene, very similar (kind of a rip-off probably) to the separatists roundtable scene in ATOC, with CG aliens speaking strange languages and everything - and i was shocked because it was a tv show, with a scene at almost (about 90%) the same quality level as the equivalent scene in ATOC, which i consider to be one of the best effects movies of all time. i think shows like enterprise are not only carrying on the legacy, but also in some ways pushing the tv domain further in terms of production quality.
are there any screenshots of 3.2 anywhere? is the look and feel any different? the kde.org site doesn't seem to have any yet.
IANAL, but my understanding is that in the U.S., and I would assume other places such as U.K. as well, theft is a criminal matter, while copyright infringement is a civil matter - two completely different areas of the law, regardless of the "similarities". I believe this is part of the reason that slashdotters often disagree with the usage of the word "theft" in conjunction with copyright law.
If the barrio (filled with illegal border-crossers) just a few blocks away from my parent's neighborhood in HOUSTON, TEXAS is still part of Houston, TX, then why aren't the residents there citizens of the United States?
clearly you have not taken the tube into the city in london at 8am on a weekday - that is most likely the type of scenario the original poster is referring to. yes, almost everyone in the train has a laptop, but no, there is virtually no room to budge, much less to pull out a 14" and prop it on your shoulder. even if you are lucky enough to get a seat, you won't be in the mood to pull out a laptop and get it powered up. a portable video device (with headphones) would be optimal for this scenario, and i'm sure a significant percentage of the thousands who do this in london every day would be interested.
actually, the "MVP"s are the ones giving the help, not getting it.
Sorry, not meaning to take it out on you specifically - far too many people are quickly throwing an analogy out like this without thinking, and you just happened to be the convenient one at the top to reply to.
;-)
Coding bugs != design flaws.
Exactly, and more importantly...
coding bugs != security vulnerabilities.
A lot of people here are making analogies about the former, while this discussion is supposed to be focused on the latter. People are confusing the two as if they were the same thing. I entirely agree that MS products have far too many bugs, but that is not relevant to whether it is insecure. Granted, the two are related, but an analogy about defects is not implicitly applicable to security, and people are using this car analogy as if it were.
Thank you, this was exactly my point. I recall watching a saturday morning special or something about this - that the architects had tried to anticipate such a thing, but not fully accomplished the level of "security" that they intended, because there were secondary issues that they had not anticipated - much like our "insecure" software. Thanks to all who got my point - the building thing may not be the perfect analogy in this case, but it is certainly better than a "car".
Is it the driver's fault for using a car that explodes when rear-ended?
It never ceases to amaze me that people continue to make this ridiculous car/software argument. the correct analogy is:
Is it the driver's fault for using a car that explodes when a rocket-propelled-grenade is launched at it?
Rear ending a car is a normal day-to-day *ACCIDENT* that normal drivers do all the time.
Writing a virus/worm/exploit to specifically attack known holes in a piece of software is not an accident, and it is not something normal users do. Show me a RPG-proof car, and then I'll build you some hacker-proof software.
Face it people: the truth is that today, software products are far more "secure" than the majority of consumer products. You are just having a hard time making the distinction between warranty failures and security attacks. We are talking about security here, not faults. I bet a lot more cars get stolen every day than OS's and email clients that get hacked.
I don't mean to be politcally distasteful here, but instead of comparing a piece of software to a car, perhaps we should be comparing it to something like a building. Both have architects, so this makes sense. An important, highly-populated building should probably be designed to anticipate and withstand a variety of terrorist attacks, such as planes flying into it. Recent events show us that this is not always the case. And that cost us lives, not just dollars! But we don't constantly sit here and hate the architects - we learn to deal with it and try to find ways to get better, knowing that we will never completely eliminate all vulnerabilities the building might have to a terrorist attack. Today I don't think anyone walks into a building assuming the building is 100% invulnerable. Or a plane. Or any other highly complex, highly targeted construction.
Making analogies to cars worked when we were complaining about defaults. But it doesn't make any sense when we are talking about security. Let's try to use some better analogies, and think about whether they are relevant.
thanks for the user page tip - i now realize that my profile is already set up for this, but unfortunately i've been receiving all my notifications on my hot[spam]mail account... i'll have to adjust that...
.NET is doing anything new.
i'm finding that at least philosophically, we have more agreement than we have disagreement - i certainly don't debate your feelings about what is "interesting" in the software world, or who is able to "accurately" predict software value - for the most part, i entirely agree.
my point about valuations is not to imply that i believe the method to arrive at those valuations is anywhere near correct or accurate - but the fact remains that while a language like C# may be nothing really new under the sun, there is still "value" in learning it for some (a lot of) developers, because those developers realize that that skillset == $$$ in their particular cases. in this case, developer income == the value or "worth" in learning the new language.
As a result, I don't see that
While the list you've given contains a number of wonderful names, I still stand by my point. The point is that 95% or more of the world's paid developers today are split between Java, VB, and C++ - and maybe Cobol and Fortran still make up a big chunk, but even then the point still stands...
Thus the "interesting" thing in this case is not a "software-interesting" thing - it is a "marketshare-interesting" thing - MS has released into the programming world the first set of tools *feasably capable* of meeting the needs of a vast majority of developers out there today.
To do this, you can't just have Java, a bunch of legacy languages, and a bunch of revolutionary new paradigm and scripting languages - you must also have the other staples of the market - thus you must have VB and C++. If for no other reason, this is because stubborn C++ guys are never going to give up, and a miserable majority of VB developers are sadly never going to make it to another environment before they have a career change.
Not the "interesting" you were hoping for? Hmm, maybe not - maybe not for me either - but for my team which may consist of a few hopeless VB and C++ guys, it makes all the difference.
I think you're on to the most important issue here, and everyone seems to be overlooking it. The biggest impact a ban like this can have is to give spammers insurance that they won't have to worry about this sort of anti-spam tax in the future. You are definitely right that that is (was?) one of the most promising schemes anyone has thought of to stop spam.
:)
*sigh* oh well, at least I won't have to pay taxes... I can feel so happy to know that the $4.27 a month I'm saving by not paying these taxes only costs me 10 hours a month of sifting through spam!
P.s. please mod this guy's comment up!
rjh, Not sure if you're still following this thread - if you have some kind of auto-email thing turned on let me know, because I'm a relative /. newbie and have no idea how to set that up if it exists...
.NET architecture with an engine based on a different par
Thanks for taking the time to continue this discussion and write such a thought-out post.
It's meant to say that in hindsight many things become obvious which required real genius to see them in the first place.
Totally agree, and in the case of graph theory, as I said, Dijkstra definitely seems to have made a significant contribution, even if many of his findings do now seem somewhat obvious. What I really wanted to point out was the fallacy of the "language worth learning" quote, in that it only uses "worth" in once sense of value, when in fact the majority of people in the world who deal with technology use a completely different set of criteria on which to place the "value" of learning a language - and this usually comes down to $ in the end.
My insistence on multiple paradigms is unequivocally not born out of "academia", a word which it seems you're using as an insult. My insistence on multiple paradigms is born out of my demand that I use the right tool for the job.
Let me apologize as I see that it probably sounds like that's the connotation I'm putting on this word. I absolutely don't mean that - I simply mean that in this particular discussion, I am talking more about industry values, because that's what seems most relevant. I am all for the academic interest in alternative paradigms, and to a certain extent I fully agree with what you're saying here - that paradigms are a part of the "right tool for the job" decision. I simply don't believe I've seen this to be the case as much in real-world scenarios. I would love to have a particular area of the projects I work on which is best suited to Lisp, and at the same time to have the appropriate toolset to do so - but today it just isn't happening. Granted - one of the previous companies I worked with was a civil engineering firm, and the architecture of their systems was all about putting the engineering/mathematical models in Fortran, and then integrating that with object-oriented and COM tools on the front end and business layers - but I honestly believe the choice for Fortran was primarily a result of (1) applicable skillsets and low budget for training/upgrading of skillsets; and (2) technical capabilities of the particular toolset (Digital Fortran) in relation to certain mathematical optimizations (i.e. optimized array/matrix calculations); certainly these are great reasons in the real-world, but in what I'm calling the "academic" sense, there is actually nothing really interesting or even nice about this particular cross-paradigm use - if they had the same skillset level and also the same level of mathematical optimizations in a Java toolset, my academic sense tells me that the applications we built would have been much more nicely architected because there is no reason those parts of the application couldn't have taken more advantage of an object-oriented environment.
Interestingly enough, on the project I am currently involved in, a major strategic part of the project is a business-level "intelligence" layer, which essentially should allow for the development of intelligent strategies to manipulate interaction with the market. Our team made an attempt to build this layer to enable the strategies to be developed fully in an alternative paradigm language. The language is called ILog, and from what I understand, it is some sort of rules-based language, essentially based on pre-conditions and post-conditions, causing certain rules to fire. Unfortunately I wasn't involved on that part of the project, so I don't know much more about it.
But suffice it to say that this is a case where I can certainly validate your claims - extreme challenges were found when trying to integrate an object-oriented
First, I'll ask you to please see my comments in the other response to you...
.NET, when people say it supports multiple languages, what they are talking about it the real-world effect of that - the idea is that if I have 2 teams with different skillsets (i.e. one with VB background, one with Java background, but both with new "upgraded" .NET skillsets), I can still have them working together in harmony, with little integration overhead, because .NET provides *this* type of cross-language integration. This whole cross-paradigm thing is extremely interesting, but given that the major market of developers .NET has to target right now are split between C++, VB, and C# - since all of those can be adapted to be OO languages, the market does not really dictate that MS must focus right now on building cross-paradigm tools.
As Dijkstra famously said, "Any language which does not change the way you think about programming is not worth learning."
Ah, classic Dijkstra. At the risk of becoming flamebait for millions of CS-Drones, let me tell you a little something about him - I had him as a professor one semester - he was not always in touch with the real world, aspecially as he got older and the world got older - apparently they seemed to take divergent paths. He may have been dead on in the 70s when he "ingeniously" realized that GOTO was a bad idea, and maybe there was a better way. In fact, even his "brilliant" graph algorithms may have been top-class during his time.
But the statement above is a classic Djikstra statement, and although I don't specifically recall it, I have no doubt he said it, because it embodies exactly who he was. If in your academic mind you truly believe this, then that's good for you - you have an interesting career ahead of you, always looking for the next paradigm. But in the real world, there are many languages in each paradigm, and sometimes the reason it is "worth" learning a new language is not for academic interest - sometimes it is productivity of toolsets supporting that language, or ability of integration with other technologies, or *shock!* money! In the real world, there are many reasons it might be "worth" learning a new language, even if it's just like another one you already know.
As it relates to
CLR and .NET are not multilanguage in any useful or interesting sense. In order to be multilanguage in a useful and/or interesting sense, it would have to support a wide variety of programming paradigms.
.NET is clearly multi-language for all practical purposes. Just because you refuse to make a distinction between "cross-language" and "cross-paradigm", it does not follow that no such distinction exists. .NET supports more than one language, hence it is multi-language. How much simpler can it be?
.NET are, let's continue this discussion.
This statement may be true in your academic mind, but in the real world,
In regards to the rest of your comments regarding more historically interesting languages, and their likelihood of sticking around for a while - I am all on board with you. Lisp has long been one of my favorite languages, primarily because for me (and I'm sure others as well) a way of recursive thinking suits me better. I certainly won't disagree with you on the academic points, nor on your points about what may remain in the distant future. However, in the real world, and in the domain where this discussion started, the here and now is what was supposed to count.
The day I can find a job working on Lisp projects as interesting (in a business sense, not an academic sense) as the projects I work on in
Yes, but in the same phrase it then says "...allows objects..."
.NET does not support "all programming languages", so any argument derived from a marketing statement based on that premise is flawed from the start.
Clearly the domain of "languages which can understand objects" is implied. furthermore, although this is not written in the most straightforward way, "all languages" is clearly meant to imply all languages in the domain of ".NET-enabled programming languages".
In any case, I don't care much for this marketing spheel. I was under the impression that we were discussing the actual real-world multi-language capabilities of the CLR. Clearly