Offtopic? The parent poster clearly stated their perception of the position of academia, in regards to Mundie's comments, to which I questioned the relevance of such a position. That's the whole idea of threads.
You're an idiot! Only someone lacking in
any cognitive abilities would make such a
sweeping comment about the academic community....It is likely that the political leanings of academics - taken as a whole - is different than non-academics, however, you have NO data to support your ignorant comment
Uh huh...need more be said? Firstly I am an "idiot" and my "generalization" (if it could be called that, though all I said was that academia is rife with socialist ideologies, which neither indicates 100%, or even that it's great than 50%: Rather that there is a greater incidence than in the general "non academia" population) is "remarkably stupid", and then you completely support my argument, which is that the political leanings of "academics" does not correlate that of the non-academic world.
The irony is that I am neither agreeing nor disagreeing with Mr. Mundie, nor did I make any assertions as such (so in what manner do I have to support an argument that I didn't even present?). Don't presume that I agree with him just because I question the finality or accuracy of the ivory tower's take on things: It just happens to be that in the comfortable environment that is academia (where bizarre notions like tenure enter the equation) that socialist concepts tend to flourish.
Ironically, their continual 'poking' and 'name calling' seems to be making it more and more popular - I dont know a single person who reads a fresh cut of Microsoft FUD and says "My god - they are so right - time to ditch this Linux crap, and buy 200 copies of Windows!".
99.99% of the regular population couldn't care less about the GPL or Microsoft's take on it, and I hardly think Everyman Joe is going to install Linux because he read that Microsoft doesn't like the GPL. Indeed, I had not read about this anywhere except for her on Slashdot, which is of course where the preaching to the choir takes place.
I don't know a single person who reads a fresh cut of Stallman/ESR/others FUD and converts to Linux/GPLd software/etc.
That really doesn't make an awful lot of sense. Does terrorism go away if you pretend it doesn't exist? (no, I'm not comparing the GPL with terrorism. It just happens to be a convenient example in today's news environment). Microsoft is paying attention to the GPL because there are countless other people advocating it, or employees subtly subterfuging it into company products : People _DO_ need to be aware of it. Whether Microsoft's take on it is correct is not what I am judging, but the fact that they make comments on it seems reasonable to me.
Having said that: Any company that touches GPLd code with a 20 foot pole needs to ferret out the zealots in their midst : How many Slashdot stories have their been now crusading against some GPL violation or another? For all of the talk about the GPL and commercial software being compatible, it is ironic seeing the countless "down with evil commercial software!" tirades on here (almost always unjust, but such details as facts elude the GPL crusaders).
In the academic community he'd be shot down and discredited so fast that his head would spin.
And it just so happens that the academic community is rife with socialist ideology (I didn't miss the Marx in your name), with countless profs who will wax poetically about how capitalism is evil, open source is salvation, and Linux is the true path to enlightenment. Whether this has any correlation with the real world is unknown.
Well in those instances it sortof is a question, with the question being "dontcha think?".:-) Seriously though, eh in Canadiana is turning a simple statement into a conversation : "It's cold out, eh?" is not just a guy making an observation about the weather, but rather soliciting conversation from the other party in the conversation. Almost every area has a variant that serves the same purpose.
Of course 95% of "Canadiana" that US media likes to mock is actually Minnesotan traits (I don't think there's anything mockworthy about any local trait, but every now and then you get that inferiority complex monkey that tries to pick on regionalisms as inferior somehow).
Actually my comment was more regarding whether it is "stupid" to not care about ads, etc: I find such elitism tiring and flawed.
In any case it absolutely "costs" to filter ads because it deprives them of income, or it undermines the entire advertising industry for the internet (which has the long term affect of depriving them of income). If a movie theater is half-full does it "cost them nothing" for you to sneak in? Sure it does if you think in a selfish "me-alone" mentality, but then why can't everyone sneak in based upon that same theory. As a whole does it "cost nothing"?
Stupidity? I'd say "low motivation to overcome it" would be a much better statement. Most people simply couldn't care less and aren't going to bother will filtering or other countermeasures. Equating that with stupidity is ridiculous.
I could get into the pay parking garage for free by having a helicopter airlift me in and out, but just because it CAN be done doesn't mean that I'll do it.
How much do you think it costs to send say 100 gig of data down the pipe? Cmon "Mr.Bandwidth doesn't grow on trees". How much? I'll tell you how much. After the hardware is paid for (which it was in the 90's for the most part) It costs fucking pennies, if that.
I highly recommend you sign up for "Economics 101". If someone puts $1 billion of hardware out there, they expect a RETURN on that $1 billion worth of hardware (if you believe that is evil then please pony up that billion yourself) that at least equates what they could get if they invested it in the general markets (i.e. at least 6%), and that's ignoring that the internet today is VASTLY changed from the infrastructure put in place in "the 90's for the most part" : Want to back up that?). Don't like it? Build your own friggin' system.
Hrm, not like anyone working for Microsoft reads Slashdot, if they did, they probably wouldn't work there for long...
Huh? Villainize Microsoft as much as you want, but as far as tech employers go Microsoft is by far among the best (I don't work for them, but by what I've heard. This great treatment of employees is one of the primary reasons why MS is where it is today). Because of that I'm often baffled by so many in the industry, working for companies that stuff you in tiny cubes and treat you as chub meat, go out with fangs out against Microsoft.
Plenty of people from Microsoft read Slashdot (and I doubt as sinister "covert operatives" either. It's (sometimes) interesting stuff whether you run Linux or not. I have FreeBSD on a firewall, but apart from that it's all Windows baby, yet I read Slashdot), provable if you've had a link on here and you see the hits come it from Redmond.
That gets into a gray area where you really have to define faulty. For instance, when it comes to system faults vendors should be required to offer a guaranteed uptime (they can set the value at whatever they want, so you could sell your software with a guarantee of no more than 20 critical faults a minute, but that might hurt your sales somewhat... As it is, organizations make very few commitments to their systems, allowing Microsoft, as an example, to simply push each new OS as "way more stable that that last piece of software which we sold you under the pretense that it was super duper stable..."). Is that bicycle fault if the rider drives irresponsibly and gets hit in traffic? Is that bicycle faulty if it gets stolen or is otherwise maliciously used?
Security robustness is a marketing function (it's a feature, if you will, just like a Volvo withstands impacts better than most other cars), and insofar as vendors don't outright lie about the security of their systems, they should not be held responsible: The responsible parties are the hackers/DOS attackers/etc, and no one should ever fool themselves into anything otherwise. For all of the talk comparing software to the "real" world, the reality is that the window maker isn't responsible if someone throws a brick through it, and the lock company isn't legally responsible if someone drives a tow truck through the door: As long as it withstood at least the marketed capabilities there is no vendor fault.
You're posting that almost two hours after the article was posted, when the Slashdot effect has subsided greatly. When the article was originally posted, I couldn't connect for numerous tries. Perhaps the mirrors should have a timeout on them.
9 times out of 10 the person posting a karma whore verbatim of an article gets modded into oblivion, but every now and then when the host server is actually having problems (as it is right now), people do appreciate being able to read what the article is about, and they mod it up.
Oh I have no doubt whatsoever that ASP.NET (and C#) will take over on IIS hosted websites : JScript and VBScript were always weak server side languages, and there was no doubt that there would be some serious improvements in them. C# and the CLR simplifies accessing the system (for all of those who used COM for cross language access to system internals it isn't that great of a leap forward though. Already most internal dev teams were COMifying their subsystems : i.e. ADSI, MSMQ, etc.).
The difference between "for your employer" and "shrinkwrap" is usually merely one of scope (otherwise are you saying that in house applications don't require the same performance/scalability as shrinkwrap?), and the reality is that large projects (which many "for your employer" apps become) do not benefit from silver bullet languages: All that benefits is a 0-demo, and that usually gives the flawed perception of greatly increased productivity.
Exactly as I stated: There are countless places where RAD languages and tools are extremely beneficial (for apps that truly are micro-apps, usually with few requirements for performance), but it is interesting how so many organizations who promote a development paradigm don't eat their own dogfood.
Visual Studio.NET includes Visual C++, and it remains the most important of the products (I'm sure it irritates the VC team seeing the marketing droids out in force selling VB.NET and C#.NET when the C++ engine in Visual Studio.NET continues to be improved, with optimization enhancements that further extend its performance brilliance).
It's funny how C++ (and the subset C) is constantly glossed over as some archaic remnant of the past, yet the overwhelming majority of commercial appliations continue to be developed in C++, and likely will continue to for some time. It's an interesting scenario of Fire and Motion. It always makes me wonder if all of the.NET and Visual Basic fanatics every stop and think "Gosh, how many Microsoft products are built with.NET or Visual Basic?" (they'd be surprized that the number is very close to 0, and will likely remain so). Silver bullet languages give short term bursts of productivity, followed by the reality that the nuances of languages become trivial in the long term of a real project.
Just intriguing to see. J2EE,.NET, etc., all definitely have a place, but it is interesting seeing how many people hop on the bandwagon without requiring the developing company to prove that they eat their own dogfood.
I suppose that is true for certain types of applications (i.e. streaming video recompression or something like that) that are completely I/O bound, but for most users the hard drive is a side effect: With 512MB I hardly ever hear the hard drive doing anything whatsoever once the applications are loaded. Indeed, hard drive benchmarks have mostly disappeared as most evaluations have found that with current caching infrastructures (which has become even more of a factor now that memory is so cheap) that the hard drives all clustered together on the performance chart, because caching was the great equalizer.
Even the most cutting edge 3D games don't use current 3D processors to their potential, these days.
What games are you playing? Firstly, of course games are usually limited by the lowest common denominator (meaning that you severely limit the polygon count if 50% of the population is using the Virge 3D), but secondly there are some games that seriously tax current hardware: An excellent example is "Operation Flashpoint", which on a GeForce 3 Ti200 has a visibly stuttering frame rate at 1024x768/32-bit colour with a reasonable set of options (I'd say that the frame rate is from 10-25 FPS), yet even that game represents a massive set of compromises: Visibility is limited to 800m or so, there is a limited number of units in a set area, land is mostly defined by textures rather than polygons as polygons are too expensive. Even in the venerable Quake 3 with the mod urban terror, some of the maps (which still represent a massive collectin of compromises) send the previously mentioned video card begging for mercy in parts (and Q3 is OLD). And 1024x768 is hardly a great resolution, and of course if you want to use FSAA you'd better knock down to 800x600. Saying that "cutting edge" games don't use the hardware to its potential makes me presume that the most demanding game you've played is The Sims (though even it can get stuttery when you have fully decked out a multi-level house, and it is hardly an example of photo-realism).
The "too much power" argument has always been flawed, going back to when the 486 was introduced and countless pundits exclaimed that a 386/33DX was all anyone needed. This same argument has gone on, foolishly, since the beginning of computers I'm sure. Actually probably back to the abacus.
Those who fail to learn from history...
on
Dot.Con
·
· Score: 2
So a post-mortem analysis on what happened, and why it happened, has no merit? As the old and incredibly wise saying goes, those who fail to learn from history are doomed to repeat it.
I don't vouch for this book whatsoever (haven't read it), but learning from the.COM bubble is extraordinarily valuable, both from a "herd mentality" investing example, and from a "rich get richer" perspective (i.e. a lot of banks and big investors made and kept an incredible amount of money when every average Joe rushed in to free them of those stocks at ridiculously inflated prices, all the while being urged on by "analysts" paraded as impartial while being employed by a company that depends upon boisterous IPOs). The whole system is grossly flawed.
There was a very good PBS special on a couple of weeks ago, and it was very enlightening (at least for me). Of course it, too, was called dot.con, much to the consternation of many on here (I think the slamming of the name is a little ridiculous. It isn't a literal domain name so semantics on whether or not it's redundant are absurd, but putting.CON makes it a tad difficult to find in the card catalog).
Liberalism is seldom associated with increased surveilance and invasions of personal liberties: Quite the opposite in fact. Most "soccer moms" who call for greater and greater restraints and government controls are conservatives. A liberal approach to things is live and let live. A conservative approach to things is "live the way we see as the best way to live".
You think so? I don't know what sites you visit, but a huge swath of great independent sites that I long visited are now mere memories (dozens upon dozens upon dozens). Instead, the net is owned by the big boys such as MSNBC, AOL, etc.
More people=more money=more infrastructure. Of course the old timers have been proclaiming "DEATH OF THE INTERNET! STORY AT 11" for years based on the same suppositions as yours.
Having said that, pardoxically with the more people, the free content has largely dried up: You can't stream those Adcritic ads any more, and download sites have clamped shut or imposed limits, and just the general amount of broadband accessible content has diminished. I'd wager that the average Internet user downloads significantly less today than they did 2 years ago.
...the online advertising revenue model collapsed, and the total content (fresh, updated content) has imploded to a fraction of what it was 2 years ago, with many sites that were vibrant not placeholders or shells.
Offtopic? The parent poster clearly stated their perception of the position of academia, in regards to Mundie's comments, to which I questioned the relevance of such a position. That's the whole idea of threads.
You're an idiot! Only someone lacking in any cognitive abilities would make such a sweeping comment about the academic community....It is likely that the political leanings of academics - taken as a whole - is different than non-academics, however, you have NO data to support your ignorant comment
Uh huh...need more be said? Firstly I am an "idiot" and my "generalization" (if it could be called that, though all I said was that academia is rife with socialist ideologies, which neither indicates 100%, or even that it's great than 50%: Rather that there is a greater incidence than in the general "non academia" population) is "remarkably stupid", and then you completely support my argument, which is that the political leanings of "academics" does not correlate that of the non-academic world.
Choose an argument before you go on a tirade.
The irony is that I am neither agreeing nor disagreeing with Mr. Mundie, nor did I make any assertions as such (so in what manner do I have to support an argument that I didn't even present?). Don't presume that I agree with him just because I question the finality or accuracy of the ivory tower's take on things: It just happens to be that in the comfortable environment that is academia (where bizarre notions like tenure enter the equation) that socialist concepts tend to flourish.
Ironically, their continual 'poking' and 'name calling' seems to be making it more and more popular - I dont know a single person who reads a fresh cut of Microsoft FUD and says "My god - they are so right - time to ditch this Linux crap, and buy 200 copies of Windows!".
99.99% of the regular population couldn't care less about the GPL or Microsoft's take on it, and I hardly think Everyman Joe is going to install Linux because he read that Microsoft doesn't like the GPL. Indeed, I had not read about this anywhere except for her on Slashdot, which is of course where the preaching to the choir takes place.
I don't know a single person who reads a fresh cut of Stallman/ESR/others FUD and converts to Linux/GPLd software/etc.
That really doesn't make an awful lot of sense. Does terrorism go away if you pretend it doesn't exist? (no, I'm not comparing the GPL with terrorism. It just happens to be a convenient example in today's news environment). Microsoft is paying attention to the GPL because there are countless other people advocating it, or employees subtly subterfuging it into company products : People _DO_ need to be aware of it. Whether Microsoft's take on it is correct is not what I am judging, but the fact that they make comments on it seems reasonable to me.
Having said that: Any company that touches GPLd code with a 20 foot pole needs to ferret out the zealots in their midst : How many Slashdot stories have their been now crusading against some GPL violation or another? For all of the talk about the GPL and commercial software being compatible, it is ironic seeing the countless "down with evil commercial software!" tirades on here (almost always unjust, but such details as facts elude the GPL crusaders).
In the academic community he'd be shot down and discredited so fast that his head would spin.
And it just so happens that the academic community is rife with socialist ideology (I didn't miss the Marx in your name), with countless profs who will wax poetically about how capitalism is evil, open source is salvation, and Linux is the true path to enlightenment. Whether this has any correlation with the real world is unknown.
Well in those instances it sortof is a question, with the question being "dontcha think?". :-) Seriously though, eh in Canadiana is turning a simple statement into a conversation : "It's cold out, eh?" is not just a guy making an observation about the weather, but rather soliciting conversation from the other party in the conversation. Almost every area has a variant that serves the same purpose.
Of course 95% of "Canadiana" that US media likes to mock is actually Minnesotan traits (I don't think there's anything mockworthy about any local trait, but every now and then you get that inferiority complex monkey that tries to pick on regionalisms as inferior somehow).
Actually my comment was more regarding whether it is "stupid" to not care about ads, etc: I find such elitism tiring and flawed.
In any case it absolutely "costs" to filter ads because it deprives them of income, or it undermines the entire advertising industry for the internet (which has the long term affect of depriving them of income). If a movie theater is half-full does it "cost them nothing" for you to sneak in? Sure it does if you think in a selfish "me-alone" mentality, but then why can't everyone sneak in based upon that same theory. As a whole does it "cost nothing"?
Stupidity? I'd say "low motivation to overcome it" would be a much better statement. Most people simply couldn't care less and aren't going to bother will filtering or other countermeasures. Equating that with stupidity is ridiculous.
I could get into the pay parking garage for free by having a helicopter airlift me in and out, but just because it CAN be done doesn't mean that I'll do it.
How much do you think it costs to send say 100 gig of data down the pipe? Cmon "Mr.Bandwidth doesn't grow on trees". How much? I'll tell you how much. After the hardware is paid for (which it was in the 90's for the most part) It costs fucking pennies, if that.
I highly recommend you sign up for "Economics 101". If someone puts $1 billion of hardware out there, they expect a RETURN on that $1 billion worth of hardware (if you believe that is evil then please pony up that billion yourself) that at least equates what they could get if they invested it in the general markets (i.e. at least 6%), and that's ignoring that the internet today is VASTLY changed from the infrastructure put in place in "the 90's for the most part" : Want to back up that?). Don't like it? Build your own friggin' system.
Hrm, not like anyone working for Microsoft reads Slashdot, if they did, they probably wouldn't work there for long...
Huh? Villainize Microsoft as much as you want, but as far as tech employers go Microsoft is by far among the best (I don't work for them, but by what I've heard. This great treatment of employees is one of the primary reasons why MS is where it is today). Because of that I'm often baffled by so many in the industry, working for companies that stuff you in tiny cubes and treat you as chub meat, go out with fangs out against Microsoft.
Plenty of people from Microsoft read Slashdot (and I doubt as sinister "covert operatives" either. It's (sometimes) interesting stuff whether you run Linux or not. I have FreeBSD on a firewall, but apart from that it's all Windows baby, yet I read Slashdot), provable if you've had a link on here and you see the hits come it from Redmond.
That gets into a gray area where you really have to define faulty. For instance, when it comes to system faults vendors should be required to offer a guaranteed uptime (they can set the value at whatever they want, so you could sell your software with a guarantee of no more than 20 critical faults a minute, but that might hurt your sales somewhat... As it is, organizations make very few commitments to their systems, allowing Microsoft, as an example, to simply push each new OS as "way more stable that that last piece of software which we sold you under the pretense that it was super duper stable..."). Is that bicycle fault if the rider drives irresponsibly and gets hit in traffic? Is that bicycle faulty if it gets stolen or is otherwise maliciously used?
Security robustness is a marketing function (it's a feature, if you will, just like a Volvo withstands impacts better than most other cars), and insofar as vendors don't outright lie about the security of their systems, they should not be held responsible: The responsible parties are the hackers/DOS attackers/etc, and no one should ever fool themselves into anything otherwise. For all of the talk comparing software to the "real" world, the reality is that the window maker isn't responsible if someone throws a brick through it, and the lock company isn't legally responsible if someone drives a tow truck through the door: As long as it withstood at least the marketed capabilities there is no vendor fault.
You're posting that almost two hours after the article was posted, when the Slashdot effect has subsided greatly. When the article was originally posted, I couldn't connect for numerous tries. Perhaps the mirrors should have a timeout on them.
9 times out of 10 the person posting a karma whore verbatim of an article gets modded into oblivion, but every now and then when the host server is actually having problems (as it is right now), people do appreciate being able to read what the article is about, and they mod it up.
Oh I have no doubt whatsoever that ASP.NET (and C#) will take over on IIS hosted websites : JScript and VBScript were always weak server side languages, and there was no doubt that there would be some serious improvements in them. C# and the CLR simplifies accessing the system (for all of those who used COM for cross language access to system internals it isn't that great of a leap forward though. Already most internal dev teams were COMifying their subsystems : i.e. ADSI, MSMQ, etc.).
The difference between "for your employer" and "shrinkwrap" is usually merely one of scope (otherwise are you saying that in house applications don't require the same performance/scalability as shrinkwrap?), and the reality is that large projects (which many "for your employer" apps become) do not benefit from silver bullet languages: All that benefits is a 0-demo, and that usually gives the flawed perception of greatly increased productivity.
Exactly as I stated: There are countless places where RAD languages and tools are extremely beneficial (for apps that truly are micro-apps, usually with few requirements for performance), but it is interesting how so many organizations who promote a development paradigm don't eat their own dogfood.
Visual Studio.NET includes Visual C++, and it remains the most important of the products (I'm sure it irritates the VC team seeing the marketing droids out in force selling VB.NET and C#.NET when the C++ engine in Visual Studio.NET continues to be improved, with optimization enhancements that further extend its performance brilliance).
Just intriguing to see. J2EE, .NET, etc., all definitely have a place, but it is interesting seeing how many people hop on the bandwagon without requiring the developing company to prove that they eat their own dogfood.
I suppose that is true for certain types of applications (i.e. streaming video recompression or something like that) that are completely I/O bound, but for most users the hard drive is a side effect: With 512MB I hardly ever hear the hard drive doing anything whatsoever once the applications are loaded. Indeed, hard drive benchmarks have mostly disappeared as most evaluations have found that with current caching infrastructures (which has become even more of a factor now that memory is so cheap) that the hard drives all clustered together on the performance chart, because caching was the great equalizer.
Even the most cutting edge 3D games don't use current 3D processors to their potential, these days.
What games are you playing? Firstly, of course games are usually limited by the lowest common denominator (meaning that you severely limit the polygon count if 50% of the population is using the Virge 3D), but secondly there are some games that seriously tax current hardware: An excellent example is "Operation Flashpoint", which on a GeForce 3 Ti200 has a visibly stuttering frame rate at 1024x768/32-bit colour with a reasonable set of options (I'd say that the frame rate is from 10-25 FPS), yet even that game represents a massive set of compromises: Visibility is limited to 800m or so, there is a limited number of units in a set area, land is mostly defined by textures rather than polygons as polygons are too expensive. Even in the venerable Quake 3 with the mod urban terror, some of the maps (which still represent a massive collectin of compromises) send the previously mentioned video card begging for mercy in parts (and Q3 is OLD). And 1024x768 is hardly a great resolution, and of course if you want to use FSAA you'd better knock down to 800x600. Saying that "cutting edge" games don't use the hardware to its potential makes me presume that the most demanding game you've played is The Sims (though even it can get stuttery when you have fully decked out a multi-level house, and it is hardly an example of photo-realism).
The "too much power" argument has always been flawed, going back to when the 486 was introduced and countless pundits exclaimed that a 386/33DX was all anyone needed. This same argument has gone on, foolishly, since the beginning of computers I'm sure. Actually probably back to the abacus.
So a post-mortem analysis on what happened, and why it happened, has no merit? As the old and incredibly wise saying goes, those who fail to learn from history are doomed to repeat it.
I don't vouch for this book whatsoever (haven't read it), but learning from the .COM bubble is extraordinarily valuable, both from a "herd mentality" investing example, and from a "rich get richer" perspective (i.e. a lot of banks and big investors made and kept an incredible amount of money when every average Joe rushed in to free them of those stocks at ridiculously inflated prices, all the while being urged on by "analysts" paraded as impartial while being employed by a company that depends upon boisterous IPOs). The whole system is grossly flawed.
There was a very good PBS special on a couple of weeks ago, and it was very enlightening (at least for me). Of course it, too, was called dot.con, much to the consternation of many on here (I think the slamming of the name is a little ridiculous. It isn't a literal domain name so semantics on whether or not it's redundant are absurd, but putting .CON makes it a tad difficult to find in the card catalog).
Liberalism is seldom associated with increased surveilance and invasions of personal liberties: Quite the opposite in fact. Most "soccer moms" who call for greater and greater restraints and government controls are conservatives. A liberal approach to things is live and let live. A conservative approach to things is "live the way we see as the best way to live".
You think so? I don't know what sites you visit, but a huge swath of great independent sites that I long visited are now mere memories (dozens upon dozens upon dozens). Instead, the net is owned by the big boys such as MSNBC, AOL, etc.
More people=more money=more infrastructure. Of course the old timers have been proclaiming "DEATH OF THE INTERNET! STORY AT 11" for years based on the same suppositions as yours. Having said that, pardoxically with the more people, the free content has largely dried up: You can't stream those Adcritic ads any more, and download sites have clamped shut or imposed limits, and just the general amount of broadband accessible content has diminished. I'd wager that the average Internet user downloads significantly less today than they did 2 years ago.
...the online advertising revenue model collapsed, and the total content (fresh, updated content) has imploded to a fraction of what it was 2 years ago, with many sites that were vibrant not placeholders or shells.