Bioinformatics: The use of computers in solving information problems in the life sciences.
This says nothing about bioinformatics being used solely for genomics, though I hear your gripe, as many think of the two as the same. No doubt, this author has made the same assumption. I speculate it has something to do with money, since genomics are a "hot topic". The point is, you may be a bioinformatician and not even know it.
In their own words they are, "The Bioperl Project is an international association of developers of open source Perl tools for bioinformatics, genomics and life science research."
There bioinformatitians can find a wealth of useful Perl scripts and modules to use in their efforts.
Yet another example of an open source initiative serving the needs of science!
Is this story a hoax?
Since publication of this story, CNN and other media have been criticized for falling for a clear hoax. According to popular technical web site slashdot.org the story is full of holes.
"Three 100 Watt light bulbs created a drain of 4500 Watts", - it should be 300 Watts. The inventor comments that perpetual motion is impossible, but then says what he's created is a "self-sustaining unit" that generates surplus energy, surely just another name for the same thing?
Slashdot points out that this inventor's claim contravenes the second law of thermodynamics which states that in a closed system, any real physical process ends with less useful energy than it started with, some is always wasted.
In other words, a perpetual motion machine is impossible.
Not the best retraction I've ever seen, but it qualifies. Ok, where is my T-Shirt?
This is probably the lamest thing I've ever heard. What is sad is that it costs money to come up with these ridiculous ideas (see dying business model and dot com crash of 2001).
What will be under the tree, URL's?
Imagine the surge of web traffic at 5:00am when the kids get on to see what they got.
And, what will they play with Christmas morning? A printout of their gift?
What does Santa have to say about this malaka?
Games, information, audio, and someday video, that is what the web gives us. It is a tool for media and communication. As for the online store, it is a simple extension of the mail order catalog. People try to sensationalize the web store as the next great marketplace, when all they really have is a new turn on the catalog. Actually, it's worse, as jpeg and gif resolution on your screen never approaches photo resolution on your catalog. Also, unless its the site for a legitimate and reputable store, you'll never know who you are dealing with. Web stores save paper and the cost of printing, but as a business model, they lack a way to get people interested in your site vs. countless others. Customers have limitless options, thereby diluting your customer base. This article is pure fantasy, and certainly not news-worthy.
How many people give you a catalog for Christmas and point you to what they got you, telling you that you can change it if you don't like it? I'll bet none.
For certain mathematical and statistical problems, computers give people a means (not an end) to learn methods of analysis far too time consuming to perform by hand. Likewise, computers give people a means of communication not possible in their absense.
Having computers in schools allows students to develop an understanding of where, when, and how to use these iterative methods. This falls under 'rithmetic.
Like all sciences, computer science has its own language. Knowledge of the English language is usually not enough if you want to read a white paper. This falls under reading.
To contribute to the science of computer science, one must know this language well enough to compose in it. This falls under writing
By your argument, physics, biology, and chemistry are also unnecessary components of the cirriculum.
I must disagree with your statement to the effect that while Free Software would be innappropriate for desktop training at schools, it would be ideal for training in Computer Science and Engineering.
When I was in high school, the main computer lab was used almost exclusively for such training, there was another lab of computers to be used for word processing, learning apps, etc.
*nix OSs would be wonderful tools for high school students to learn to administer their computer networks or learn to write useful, reusable, and platform-independent code from the start (given the ease of linking processes in *nix style OSs, I could imagine higher level students working on components of a real program, not just 'hello world').
We have high school students working on second semester college calculus, but we are afraid they won't understand how to use a hash table, nor do we concieve that they could apply their knowledge of matrix math to interesting problems in computer science (teach them OpenGL, I say, that would spark interest, I guarantee). Instead, teachers of CS try to use windowing to get students involved and interested. After all, then the students get to see their work as a "real program", right? Such a tactic is not in the best interest of the science. Most of the real work in a program is done beneath the hood; simply teaching kids to write GUI code is a diservice to them and to the science.
Many will say the slow pace of CS instruction is a condition of the lack of qualified instructors and this is perhaps true (if you don't think the pace is slow, then explain to me why it took our school three years at an hour a day to get to linked lists? All of us geeks were so beyond the level of instruction that class became a second study hall) However, while computers may confound a good portion of the masses today, this will not be the case in the future. As the science matures, general knowledge of the subject will spread and qualified instructors will be more abundant.
If schools are resigned to teaching kids using Microsoft Visual Studio, they are going to suffer in not understanding the underlying mechanisms of windowing, figure a wizard will always write that stuff anyway, and be subject to Microsoft's idea of the best architecture for a program (Document/View?, don't get me started).
So you say, well, you can do all of things I promote on a windows box, can't you? Sure, and as long as you don't need to pipe between more than two processes, you shouldn't have too much trouble. My point is, why spend the money if you don't have to? School budgets are tight, and monies would be much better spent on competent instructors than idiot-proof windows boxes and licenses. Limiting the machines in a teaching lab to windows doesn't make sense. Old windows machines can be recycled by loading them with free *nix-style OSs, at no cost. We do that here at work; a 133MHz machine is suddenly useable again! (Here's a thought, teach HS students to build a cluster out of "obsolete" machines). Also, with only windows boxes available, will students appreciate the reason open standards help forward the science, or will Microsoft be responsible for forwarding the science?
IMHO, CS classes should be paced with mathematics classes at the high school level. Teach our kids critical analysis skills and emphasize the power of the concepts taught them. Most important, give them the proper tools for the job, and if one of those tools is available at no cost, then using it is a no-brainer.
Online games use protection. Diablo II checks the disk serial number when you sign in to BattleNet, preventing simultaneous use from an original and copied CD. So far as I know, this strategy has worked, but it could only be used in an online-type of situation. That is why the record industry needs the cooperation of the OS, entre Microsoft
A question to/.'ers with knowledge of the copy protection scheme to be used...
I have read very little on the technical nature of the topic, probably because such a hyped issue inevitably lands in hands of the McPress. Most of what I have seen though, explains such copy protection as introducing "cracks or irritating noise" into the music, most of which is filtered out by today's noise reduction systems common in CD players, but not in computer CD drives.
So, my question is, could this noise be removed by passing the music data through a nicely tuned Fast Fourier Transform?
If so, a copy/processing utility could be written in under an hour that would give you access to the sound info, albeit ripping time would certainly increase.
There are far more problems with the desktop metaphor than this article indicates. For one, I have a problem with its lack of solutions to the problems it cites. For two, I have a problem with its solutions it does give.
The article suggests that we should have several named desktops to catagorize our stuff. I can see it now, "Where the hell did I put my email from 12-18-01?". To solve the inevitable problem of people who are by their nature bad organizers, we could force them to put code on one desktop, email on another desktop, programs on another desktop, pictures of the grandkids on another desktop. This is beginning to sound like more of a burden than a benefit to me, at least, and I doubt that Windows would get it right. Is there anybody out there who really uses the "My Documents" folder for all their documents? Who was the genius who thought it should be permanent on your desktop? I think what these authors are leading us to is a more structured file system (ala *nix), because apparantly, Windows, "do as you please with your files" system is confusing to some people.
As an aside (and mostly offtopic), I see a problem with the way applications are contained by the MDI system Microsoft tries to force us into (if you use Visual Studio wizards, that is). I have a good amount of screen space, why should my program have an all-encompassing main form surrounding all the children. Is maximizing a window always necessary? Would my feeble mind be otherwise distracted by items showing on my desktop? This dead space obscures other apps I may want to have running on my desktop. On a real desktop, I don't have to put my phone in the drawer if I want to use the calculator.
I have a great idea, let's do away with the desktop and use virtual terminals.
In reading the posted article and in reviewing some literature concerning WEP security here: CS at Berkeley I was wondering if anyone out there had insight on the nature of the modifications that have been made.
Please excuse my naivety in the field, but from the Berkeley article I gather that not only is the similarity of the packet keys a weakness of WEP (as RSA indicates), but also the use of a 24-bit space for the initialization vectors used to generate the RC4 packet keys.
Now, is the 24-bit space limitation what RSA means by, "similarity of the packet keys", or are they referring to the fact that most boards start the IV at 0 and simply increment for each packet (the end result being numerous IV collisions)?
The reason I wonder is because theoretically, at least, one could construct a table of all IV + key stream combinations in a decryption table (~15Gb according to Berkeley) and thereby gain himself the key to the city, so to speak. So, while limiting the number of IV collisions would certainly make decryption more difficult and certainly more time consuming, it wouldn't make WEP entirely secure. In the event that someone be so determined to monitor WLAN activity for enough time to construct such a table, could users of WEP be exposed?
We used piezoelectric crystals in a Genomics Lab at the University of Wisconsin and there are two problems that immediately come to mind at the mention of this application (and an afterthought).
For one, they are very costly. Perhaps with their proliferation, the costs would go down, perhaps not.
Second, piezoelectric crystals are very fragile. They have a tendency to crumble when too much force is applied to them. Unless this problem has been solved, transport of such a device could easily cause damage. See point #1.
Afterthought, there may also be a problem with condensation associated with the use of piezoelectrics. Without the air flow of a fan, devices of this sort are subject to water vapor condensation, which, as everybody knows, is a bad thing to have happening on your mother board. In an analogous situation, my brother and I tried using Peltier junctions to cool our hot rod, and the result was a watery mess. (Coincidentally, I now work for an unnamed company that relies on Peltier junctions for rapid thermal cycling, and to solve the condensation problem, we have relied on, you've got it, heat sinks and fans).
Right, and if you take the development of such an OS to be a predictor of ratification, then I see reason for silent alarms. If you conclude conspiracy, well... I'd say there is a fine line between conspiracy theorists and whacked sensationalists.
Still, it is no secret that M$ is more than willing to comply with whatever is required if it means greater market share (to acheive 99.999%, the only attainable 5 9's for M$). In this case, however, I believe it is the SSSCA and not M$ that freedom-loving citizens should focus on. After all, complying with a federal mandate is just good business, and if it be a mandate on the horizon, then it is wise business to get a lead on the competition.
However, I compelled to accept the writing on the wall argument.
I don't see what the big hubbub is. If you don't like the OS, then you don't have to use it.
So you say, yeah, but Monopoly this, monopoly that, everybody develops for the monopoly, blah blah. My fellow Americans, if nobody buys it, the monopoly will stop selling it. After all, Big Blue was once seen as the oppressive monopoly, wasn't it?
The more you tighten your grip, the more star systems will slip through your fingers [Princess Leia]
Those stories covered the initial expedition, as stated in the BBC article. The news is the confirmation of the initial observations, which came on Thursday. I know, I know, so what?, but many of these expeditions are covered this way.
As to whether or not it should be on SlashDot...
...maybe if we could look at some of these images of which they speak.
Yeah, those are IDE requirements, as evidenced by the fact that you need Windows 2000 or better as an OS, which wouldn't really make sense any other way. Also, requiring a DVD drive on an embedded system... no I don't think so.
All I've seen is silly press releases and such. I noticed on the MS website, that there is a trial version of the EmbXP IDE. Has anyone experimented with this?
Else, are there any MS employees or other folks who can point to some technical information, like memory req's.?
That's funny. I looked briefly, but I couldn't find any technical info, only drum-beating. That's a good joke, but I was really wondering about the requirements of EmbXP like memory, etc. Does anybody out there have some technical info to share (links?), or is this news just for yucks?
Who are you, the fucking grammar police? Half of the fucking article is about software techniques for handling requests expeditiously to avoid network saturation, dickhead. Apparantly, these techniques were not sufficient. The article indicates a need to serve "millions of requests a day"; it fell short. If you think I meant the server ceased functioning, then you're thinking too hard, genius.
This is a pretty good method, and in fact it is the one we used to use. However, you run into problems if you want to port your code to any other platform. For us, there was a push for a Mac version, and porting VB and Active X to Mac is no picnic.
Instead of using ATL COM or Win32 Dll's, we used C++ ActiveX objects. It is nice to do this, because ActiveX objects in C++ can fire events back to VB, giving you extra control from the C++ side (in our experience, tis better to store object data and perform intensive file I/0 from C++, only GUI handlers exist in our VB). Separating the code like this has made it easier to port, since much of the C++ code was reusable. Also, you get pointers to use in your objects, which can be very nice (or evil).;)
Also, as you probably know (but if you don't), using C++ Dll's you can create a class that is child to CCmdTarget (MFC, allows automation calls) and make it a dispatchable object from the VS wizard. This allows you to create the object directly in VB (as you would any object in VB) and make function calls to it across the automation boundary, without going through the hassel of importing every Dll function in VB. This method doesn't allow events to be fired from the C++ object though; however, the object doesn't need to be placed on your form and declared invisible a la ActiveX objects.
Slashdot Mirror
From Gray's Lab Dictionary on medical sciences:
Bioinformatics: The use of computers in solving information problems in the life sciences.
This says nothing about bioinformatics being used solely for genomics, though I hear your gripe, as many think of the two as the same. No doubt, this author has made the same assumption. I speculate it has something to do with money, since genomics are a "hot topic". The point is, you may be a bioinformatician and not even know it.
The BioPerl project (http://bio.perl.org/) has been going on for some time.
In their own words they are, "The Bioperl Project is an international association of developers of open source Perl tools for bioinformatics, genomics and life science research."
There bioinformatitians can find a wealth of useful Perl scripts and modules to use in their efforts.
Yet another example of an open source initiative serving the needs of science!
from cnn.com story:
Is this story a hoax?
Since publication of this story, CNN and other media have been criticized for falling for a clear hoax. According to popular technical web site slashdot.org the story is full of holes.
"Three 100 Watt light bulbs created a drain of 4500 Watts", - it should be 300 Watts. The inventor comments that perpetual motion is impossible, but then says what he's created is a "self-sustaining unit" that generates surplus energy, surely just another name for the same thing?
Slashdot points out that this inventor's claim contravenes the second law of thermodynamics which states that in a closed system, any real physical process ends with less useful energy than it started with, some is always wasted.
In other words, a perpetual motion machine is impossible.
Not the best retraction I've ever seen, but it qualifies. Ok, where is my T-Shirt?
This is probably the lamest thing I've ever heard. What is sad is that it costs money to come up with these ridiculous ideas (see dying business model and dot com crash of 2001).
What will be under the tree, URL's?
Imagine the surge of web traffic at 5:00am when the kids get on to see what they got.
And, what will they play with Christmas morning? A printout of their gift?
What does Santa have to say about this malaka?
Games, information, audio, and someday video, that is what the web gives us. It is a tool for media and communication. As for the online store, it is a simple extension of the mail order catalog. People try to sensationalize the web store as the next great marketplace, when all they really have is a new turn on the catalog. Actually, it's worse, as jpeg and gif resolution on your screen never approaches photo resolution on your catalog. Also, unless its the site for a legitimate and reputable store, you'll never know who you are dealing with. Web stores save paper and the cost of printing, but as a business model, they lack a way to get people interested in your site vs. countless others. Customers have limitless options, thereby diluting your customer base. This article is pure fantasy, and certainly not news-worthy.
How many people give you a catalog for Christmas and point you to what they got you, telling you that you can change it if you don't like it? I'll bet none.
The only hope is to have pay channels like HBO - I think thats the only real HDTV you will see.
Not so, PBS is broadcast in High Definition in many areas for many of their programs (like Nova).
Funny, how did they get my DNA without my knowledge!?
1% of a very large number is still a very large number
... of Analysis and Communications.
For certain mathematical and statistical problems, computers give people a means (not an end) to learn methods of analysis far too time consuming to perform by hand. Likewise, computers give people a means of communication not possible in their absense.
Having computers in schools allows students to develop an understanding of where, when, and how to use these iterative methods. This falls under 'rithmetic.
Like all sciences, computer science has its own language. Knowledge of the English language is usually not enough if you want to read a white paper. This falls under reading.
To contribute to the science of computer science, one must know this language well enough to compose in it. This falls under writing
By your argument, physics, biology, and chemistry are also unnecessary components of the cirriculum.
Good point, instruction is expensive.
A good source of boxes is from recycling machines that can't run windows anymore.
Sorry, maintaining a good machine requires less time than patching for viruses every day.
I must disagree with your statement to the effect that while Free Software would be innappropriate for desktop training at schools, it would be ideal for training in Computer Science and Engineering.
When I was in high school, the main computer lab was used almost exclusively for such training, there was another lab of computers to be used for word processing, learning apps, etc.
*nix OSs would be wonderful tools for high school students to learn to administer their computer networks or learn to write useful, reusable, and platform-independent code from the start (given the ease of linking processes in *nix style OSs, I could imagine higher level students working on components of a real program, not just 'hello world').
We have high school students working on second semester college calculus, but we are afraid they won't understand how to use a hash table, nor do we concieve that they could apply their knowledge of matrix math to interesting problems in computer science (teach them OpenGL, I say, that would spark interest, I guarantee). Instead, teachers of CS try to use windowing to get students involved and interested. After all, then the students get to see their work as a "real program", right? Such a tactic is not in the best interest of the science. Most of the real work in a program is done beneath the hood; simply teaching kids to write GUI code is a diservice to them and to the science.
Many will say the slow pace of CS instruction is a condition of the lack of qualified instructors and this is perhaps true (if you don't think the pace is slow, then explain to me why it took our school three years at an hour a day to get to linked lists? All of us geeks were so beyond the level of instruction that class became a second study hall) However, while computers may confound a good portion of the masses today, this will not be the case in the future. As the science matures, general knowledge of the subject will spread and qualified instructors will be more abundant.
If schools are resigned to teaching kids using Microsoft Visual Studio, they are going to suffer in not understanding the underlying mechanisms of windowing, figure a wizard will always write that stuff anyway, and be subject to Microsoft's idea of the best architecture for a program (Document/View?, don't get me started).
So you say, well, you can do all of things I promote on a windows box, can't you? Sure, and as long as you don't need to pipe between more than two processes, you shouldn't have too much trouble. My point is, why spend the money if you don't have to? School budgets are tight, and monies would be much better spent on competent instructors than idiot-proof windows boxes and licenses. Limiting the machines in a teaching lab to windows doesn't make sense. Old windows machines can be recycled by loading them with free *nix-style OSs, at no cost. We do that here at work; a 133MHz machine is suddenly useable again! (Here's a thought, teach HS students to build a cluster out of "obsolete" machines). Also, with only windows boxes available, will students appreciate the reason open standards help forward the science, or will Microsoft be responsible for forwarding the science?
IMHO, CS classes should be paced with mathematics classes at the high school level. Teach our kids critical analysis skills and emphasize the power of the concepts taught them. Most important, give them the proper tools for the job, and if one of those tools is available at no cost, then using it is a no-brainer.
Online games use protection. Diablo II checks the disk serial number when you sign in to BattleNet, preventing simultaneous use from an original and copied CD. So far as I know, this strategy has worked, but it could only be used in an online-type of situation. That is why the record industry needs the cooperation of the OS, entre Microsoft
A question to /.'ers with knowledge of the copy protection scheme to be used...
I have read very little on the technical nature of the topic, probably because such a hyped issue inevitably lands in hands of the McPress. Most of what I have seen though, explains such copy protection as introducing "cracks or irritating noise" into the music, most of which is filtered out by today's noise reduction systems common in CD players, but not in computer CD drives.
So, my question is, could this noise be removed by passing the music data through a nicely tuned Fast Fourier Transform?
If so, a copy/processing utility could be written in under an hour that would give you access to the sound info, albeit ripping time would certainly increase.
There are far more problems with the desktop metaphor than this article indicates. For one, I have a problem with its lack of solutions to the problems it cites. For two, I have a problem with its solutions it does give.
The article suggests that we should have several named desktops to catagorize our stuff. I can see it now, "Where the hell did I put my email from 12-18-01?". To solve the inevitable problem of people who are by their nature bad organizers, we could force them to put code on one desktop, email on another desktop, programs on another desktop, pictures of the grandkids on another desktop. This is beginning to sound like more of a burden than a benefit to me, at least, and I doubt that Windows would get it right. Is there anybody out there who really uses the "My Documents" folder for all their documents? Who was the genius who thought it should be permanent on your desktop? I think what these authors are leading us to is a more structured file system (ala *nix), because apparantly, Windows, "do as you please with your files" system is confusing to some people.
As an aside (and mostly offtopic), I see a problem with the way applications are contained by the MDI system Microsoft tries to force us into (if you use Visual Studio wizards, that is). I have a good amount of screen space, why should my program have an all-encompassing main form surrounding all the children. Is maximizing a window always necessary? Would my feeble mind be otherwise distracted by items showing on my desktop? This dead space obscures other apps I may want to have running on my desktop. On a real desktop, I don't have to put my phone in the drawer if I want to use the calculator.
I have a great idea, let's do away with the desktop and use virtual terminals.
In reading the posted article and in reviewing some literature concerning WEP security here: CS at Berkeley I was wondering if anyone out there had insight on the nature of the modifications that have been made.
Please excuse my naivety in the field, but from the Berkeley article I gather that not only is the similarity of the packet keys a weakness of WEP (as RSA indicates), but also the use of a 24-bit space for the initialization vectors used to generate the RC4 packet keys.
Now, is the 24-bit space limitation what RSA means by, "similarity of the packet keys", or are they referring to the fact that most boards start the IV at 0 and simply increment for each packet (the end result being numerous IV collisions)?
The reason I wonder is because theoretically, at least, one could construct a table of all IV + key stream combinations in a decryption table (~15Gb according to Berkeley) and thereby gain himself the key to the city, so to speak. So, while limiting the number of IV collisions would certainly make decryption more difficult and certainly more time consuming, it wouldn't make WEP entirely secure. In the event that someone be so determined to monitor WLAN activity for enough time to construct such a table, could users of WEP be exposed?
A case filter will not prevent water vapor from condensing on such a device.
We used piezoelectric crystals in a Genomics Lab at the University of Wisconsin and there are two problems that immediately come to mind at the mention of this application (and an afterthought).
For one, they are very costly. Perhaps with their proliferation, the costs would go down, perhaps not.
Second, piezoelectric crystals are very fragile. They have a tendency to crumble when too much force is applied to them. Unless this problem has been solved, transport of such a device could easily cause damage. See point #1.
Afterthought, there may also be a problem with condensation associated with the use of piezoelectrics. Without the air flow of a fan, devices of this sort are subject to water vapor condensation, which, as everybody knows, is a bad thing to have happening on your mother board. In an analogous situation, my brother and I tried using Peltier junctions to cool our hot rod, and the result was a watery mess. (Coincidentally, I now work for an unnamed company that relies on Peltier junctions for rapid thermal cycling, and to solve the condensation problem, we have relied on, you've got it, heat sinks and fans).
Right, and if you take the development of such an OS to be a predictor of ratification, then I see reason for silent alarms. If you conclude conspiracy, well... I'd say there is a fine line between conspiracy theorists and whacked sensationalists.
Still, it is no secret that M$ is more than willing to comply with whatever is required if it means greater market share (to acheive 99.999%, the only attainable 5 9's for M$). In this case, however, I believe it is the SSSCA and not M$ that freedom-loving citizens should focus on. After all, complying with a federal mandate is just good business, and if it be a mandate on the horizon, then it is wise business to get a lead on the competition.
However, I compelled to accept the writing on the wall argument.
I don't see what the big hubbub is. If you don't like the OS, then you don't have to use it.
So you say, yeah, but Monopoly this, monopoly that, everybody develops for the monopoly, blah blah. My fellow Americans, if nobody buys it, the monopoly will stop selling it. After all, Big Blue was once seen as the oppressive monopoly, wasn't it?
The more you tighten your grip, the more star systems will slip through your fingers [Princess Leia]
Those stories covered the initial expedition, as stated in the BBC article. The news is the confirmation of the initial observations, which came on Thursday. I know, I know, so what?, but many of these expeditions are covered this way.
As to whether or not it should be on SlashDot...
...maybe if we could look at some of these images of which they speak.
Yeah, those are IDE requirements, as evidenced by the fact that you need Windows 2000 or better as an OS, which wouldn't really make sense any other way. Also, requiring a DVD drive on an embedded system... no I don't think so.
Obviously, but I stand that it is my Grammar that was wrong. What I meant to say was
"New Webserver For 21st Century" Goes Down
in reference to the website going down, not the server itself. Sorry about the tirade, I am easily offended.
We must end this destructive conflict and bring order to the galaxy [Darth Vader]
All I've seen is silly press releases and such. I noticed on the MS website, that there is a trial version of the EmbXP IDE. Has anyone experimented with this?
Else, are there any MS employees or other folks who can point to some technical information, like memory req's.?
We need info, not banter.
To discourage is bad karma
That's funny. I looked briefly, but I couldn't find any technical info, only drum-beating. That's a good joke, but I was really wondering about the requirements of EmbXP like memory, etc. Does anybody out there have some technical info to share (links?), or is this news just for yucks?
Who are you, the fucking grammar police? Half of the fucking article is about software techniques for handling requests expeditiously to avoid network saturation, dickhead. Apparantly, these techniques were not sufficient. The article indicates a need to serve "millions of requests a day"; it fell short. If you think I meant the server ceased functioning, then you're thinking too hard, genius.
I was enjoying the article until it /.'ed, and I couldn't get anymore pages to load.
Therein, a stress test to the folks at Ace's Hardware.
This is a pretty good method, and in fact it is the one we used to use. However, you run into problems if you want to port your code to any other platform. For us, there was a push for a Mac version, and porting VB and Active X to Mac is no picnic.
;)
Instead of using ATL COM or Win32 Dll's, we used C++ ActiveX objects. It is nice to do this, because ActiveX objects in C++ can fire events back to VB, giving you extra control from the C++ side (in our experience, tis better to store object data and perform intensive file I/0 from C++, only GUI handlers exist in our VB). Separating the code like this has made it easier to port, since much of the C++ code was reusable. Also, you get pointers to use in your objects, which can be very nice (or evil).
Also, as you probably know (but if you don't), using C++ Dll's you can create a class that is child to CCmdTarget (MFC, allows automation calls) and make it a dispatchable object from the VS wizard. This allows you to create the object directly in VB (as you would any object in VB) and make function calls to it across the automation boundary, without going through the hassel of importing every Dll function in VB. This method doesn't allow events to be fired from the C++ object though; however, the object doesn't need to be placed on your form and declared invisible a la ActiveX objects.