I'm curious - how many use psyco (another python performance enhancement tool) for their projects?
<aol>me too!</aol> Much of my code spends the majority of its time waiting for database queries to finish, ImageMagick to finish doing its stuff, files to copy, etc. Psyco doesn't do a thing for that stuff. On the other hand, a small amount of my code is pretty CPU intensive - not enough to break out of pure Python into Pyrex or anything else, but enough to want a performance upgrade. For those modules, psyco is the clear choice since I can point it at pre-existing code and get a nice boost for free.
You're freakin' high. VFP is a hugely broken piece of junk that freaks out with more than a few gigs of data and more than a couple tens of users accessing it at once. Since its "database" is made of flat text files that you share with each and every user via a fileserver, forget accessing it over anything less than a 100Mbit LAN.
My company's spending a lot of time and money migrating away from that hideous trashheap. Gentle readers, I assure you this: the only people who would ever actually recommend FoxPro are FoxPro consultants. Anyone else in the know would run screaming from the prospect.
What's actually funny, is how many people in security sensitive environments rely on those closed-source video drivers.
Well, we all have our limits. Some people worry, and justifiably so, that their BIOS isn't open. I'm somewhere in the middle in that I use the proprietary NVidia drivers, even though I don't like it. I'd think that everyone, though, would agree that the crypto engine is the absolute last thing you want to cede control of.
You got that backward. In reality, it's more like:
"Our ecommerce server is really getting slammed, and I think we should spend some money on hardware acceleration. Here's a list of prices and part numbers that are compatible with our security infrastructure."
See the difference? This isn't the kind of stuff you buy to be cool. It's the sort of thing you buy because you want to increase your server throughput. You wouldn't switch from FC3 to FreeBSD because you just have to get your hands on a particular RAID card, would you?
I can't help but think that engaging in hyperbole (50 questions? ~25 is accurate) and verbally abusing and threatening the vendor is going to help in any way.
True, but on the other hand, Theo really does have the upper hand on this one. If I can't use those cards under OpenBSD, I won't buy them. If I can, I probably will (because I could actually use some of that functionality in my VPN servers). Since I suspect a large part of their potential client base is in the same situation, it'd be in their best financial interests to go meet Theo's (reasonable) requests and stay stop arguing the point.
Hifn reserves the right to keep our source code proprietary.
Fair enough, Hank. But I reserve the right to not use proprietary crypto code in sensitive applications - which are the only ones that I'd actually buy hardware acceleration for in the first place.
Let's get this straight: there's a world of difference between closed video card drivers and closed crypto drivers. Many of us are squeamish about about the former, so why would you think we'd cheerfully accept the latter? A closed source video driver could potentially crash my non-networked game machine. A closed source encryption accelerator cold potentially open my VPN server to the whole world.
I hope you can appreciate the community's position here, but whether you agree with it or not is immaterial. Should you change your opinion to better mesh with that of your would-be customers, please let us know. Many of us would like to buy your products if they become usable for our applications.
Sure, if you're writing apps that are fine slow, interpreted doesn't matter.
Assuming, of course, that "interpreted == slow" (and that bytecode language are "interpreted") - which is demonstrably false. The trick is to let the language do as much as possible. For example, writing mylist.sort() in Python is infinitely preferable to writing your own list sorting function (in Python) and using that instead. Similarly, using map() to apply a function to each value in a list is almost always going to be faster than manually iterating across it.
Again, let a language do what that language does best: implement itself. Learning what it's capable of and taking advantage of that will give you reasonable performance in all but extreme cases.
Yeah, because all the BSDs and Linux have identical kernel interfaces, PCI subsystems, DMA handlers, etc. A simple./configure; make install is all that separates OpenBSD's kernel from 2.6.15, after all.
In reality, on the other hand, the reverse engineered drivers can serve as excellent documentation for how the same logic can be implemented in another OS, but that's about as close as it's likely to get.
On the other hand, my 2001-vintage PS2 still works flawlessly. See, that's the problem with anecdotal data - you can't extrapolate your own experience to everyone else.
What if some day the Chinese decided that they're not going to produce devices that don't meet their standards?
Then world governments dictate that all WAPI-enabled router imports ship with an OpenVPN installer CD, and we all go the sane route of running trusted VPN software over untrusted open Wi-Fi connections.
Drugs are no substitute for reading a lot, tinkering, listening to others and keeping classifying things with respect to what you already know.
True, but I could see them for being an attractive alternative to actually paying attention in "FLUFF101: Stuff Too Boring To Take Voluntarily" when you just want to learn the test well enough to get your A and get back to hacking Lisp.
I do. I never could get Reader Rabbit or Jumpstart Preschool running under Win2K, and I didn't want to buy (or pirate) XP when 98 runs my kids' games without problems.
OK, I've got a question for you: do these people actually live human lives? That is, would you bump into Darl at the mall or at a restaurant? I ask because I simply can't imagine going out of my way to make a good chunk of the population actively hate me, and another large group think I'm the walking epitome of "jackass", and then going on about my normal life.
Doesn't he have neighbors? People that fix his computer? Geeks who work at Pizza Hut spitting in his pizza? Pool cleaners who wizz in the hot tub when he's not looking? I have no desire to turn myself into a globally recognized pariah.
No, you're not. I'm still running 98 on my kids' PC because Reader Rabbit etc. don't work until Win2K and I don't want to shell out for XP for no reason. I also keep an up-to-date copy of Firefox on it so they can browse Playhouse Disney and other preschooler-friendly sites.
It's on the same LAN as my wife's relatively new iMac, a FreeBSD server, a Linux laptop, and an OpenBSD firewall. It's not that we're technically illiterate or poor, but that there's no legitimate need for us to upgrade the little gaming machine to something newer.
To the surprise of the reseachers, many adults continue to live in that mindset, and never evolve to a more nuanced ethical view in which to look at the world.
Surprise, surprise: half the population has a below-average IQ. That's extremely relevant here because working through Kant's categorical imperative isn't something something for the faint of heart (or mind). If a researcher were naive enough to be surprised that people without the intellectual ability to also become researchers didn't operate in the same mental framework, then he should have his PhD revoked.
But you're right - you don't have to follow the law. You can still expect to be held accountable for not doing so, though, even if you disagree with it.
For a different example, a Randian Objectivist sees workplace regulations as inherently immoral. Not inconvient, not annoying, but immoral. And yet, a copy of "The Fountainhead" won't get him the hook for a failed OSHA inspection. You and he can both be morally satisfied that you followed your ethics as you pay off the fines.
On the other hand, non-fundamentalists like the rest of us are puzzled at how much time the Debian team wastes in throwing tantrums over license purity instead of shipping their next version.
You're only puzzled because you don't "get" open distributions like Debian. There are thousands of people working on it at any given time, from kernel packagers to people watching the bug tracker. Among that crowd, quite a few of them are passionate about legal issues and qualified to argue them. You're hearing those people. Everyone else is still going about their business.
This is in much the same way that discussions inside a company's legal department don't involve the programming department until a relevant decision is made. It's not like the programmers have to sit on their hands until the debates are finished, and probably don't even know they're happening at any given time.
Doctors, especially in the US, have money. They can easily cover a fine of say, $200. It doesn't increase their costs greatly, but at least it does give the impression that the laws are to be taken somewhat seriously, and will be enforced.
Yeah? When does that kick in? My wife's a surgeon, but I'm stuck driving a used Oldsmobile and lately riding a bike to work. We're not poor but neither do we live up to the "rich doctor" bullshit myth that you're buying into.
In reality, doctors spend the first 12 years after high school incurring enormous debts (you didn't think med school was free, did you?), then another decade in jobs that put food on the table but just cover a mortgage and student loans. Maybe we'll be wealthy some day - if only we can start turning a real personal profit by the time we're in our 40s.
So kiss my ass with the "doctors can afford it!" BS. Some can, but a whole lot of them can't. You might as well say "IT staffers have money", or "salesmen have money", or "accountants have money"; any of those would be just as accurate (and grossly overgeneralized).
That depends on your interpretation. By using names, you're theoretically announcing that Mrs. Smith has some unknown medical condition worthy of seeing a doctor. I think it's stupid, but I've been told exactly that by a compliance auditor.
So why doesn't your wife call for number 352 from the lobby, and then ask Mr. Smith how he is once he is out of the public waiting room?
Because her patients expect to be treated like humans and humans don't like being called by number. In her practice, patients tend to be retirement age or older (except for the occasional younger person with a broken toe, etc.). That population would not react well to being numerically processed.
She's much better off business-wise to upset the one person who's not used to the standard procedure at every doctor's office they've ever been to, than upsetting the 99.99% of her patients who like being recognized as a person with a name. You can argue about whether a practice should be run as a business, but if she drives off 50% of her clientele, then she won't be able to serve the other 50% at all.
Man, I'd love to see you experience the "personal inconvenience" of being shot in the head.
He's got a point, though. That was an awful thing to have to endure, but that doesn't give Brady the right to lobby against my and your constitutional protections.
According to HIPAA, at least as of a couple years ago, no privacy violation was too small. Including, say, a nurse coming to the waiting room and asking for "Mrs. Smith". After all, Mr. Jones sitting next to her would then know that woman's name. Instead, the only proper method for calling patients back to the treatment rooms is installing one of those "take a number" dispensers, then calling patients by number.
Never mind that we live in a small town where Mrs. Smith and Mr. Jones went to kindergarten together and come from families that have been here for 150 years. And forget that my wife is a podiatrist and that visiting her isn't inherently compromising (unlike, say, sitting in the lobby of a clinic for sexually transmitted diseases).
So, according to HIPAA, my wife is breaking the law each and every time she treats her patients like people instead of numbers. We haven't had a complaint yet and don't expect to, but could technically be busted for violating Mrs. Jones's privacy at any moment.
Slashdot Mirror
<aol>me too!</aol> Much of my code spends the majority of its time waiting for database queries to finish, ImageMagick to finish doing its stuff, files to copy, etc. Psyco doesn't do a thing for that stuff. On the other hand, a small amount of my code is pretty CPU intensive - not enough to break out of pure Python into Pyrex or anything else, but enough to want a performance upgrade. For those modules, psyco is the clear choice since I can point it at pre-existing code and get a nice boost for free.
My company's spending a lot of time and money migrating away from that hideous trashheap. Gentle readers, I assure you this: the only people who would ever actually recommend FoxPro are FoxPro consultants. Anyone else in the know would run screaming from the prospect.
Well, we all have our limits. Some people worry, and justifiably so, that their BIOS isn't open. I'm somewhere in the middle in that I use the proprietary NVidia drivers, even though I don't like it. I'd think that everyone, though, would agree that the crypto engine is the absolute last thing you want to cede control of.
You got that backward. In reality, it's more like:
"Our ecommerce server is really getting slammed, and I think we should spend some money on hardware acceleration. Here's a list of prices and part numbers that are compatible with our security infrastructure."
See the difference? This isn't the kind of stuff you buy to be cool. It's the sort of thing you buy because you want to increase your server throughput. You wouldn't switch from FC3 to FreeBSD because you just have to get your hands on a particular RAID card, would you?
True, but on the other hand, Theo really does have the upper hand on this one. If I can't use those cards under OpenBSD, I won't buy them. If I can, I probably will (because I could actually use some of that functionality in my VPN servers). Since I suspect a large part of their potential client base is in the same situation, it'd be in their best financial interests to go meet Theo's (reasonable) requests and stay stop arguing the point.
Fair enough, Hank. But I reserve the right to not use proprietary crypto code in sensitive applications - which are the only ones that I'd actually buy hardware acceleration for in the first place.
Let's get this straight: there's a world of difference between closed video card drivers and closed crypto drivers. Many of us are squeamish about about the former, so why would you think we'd cheerfully accept the latter? A closed source video driver could potentially crash my non-networked game machine. A closed source encryption accelerator cold potentially open my VPN server to the whole world.
I hope you can appreciate the community's position here, but whether you agree with it or not is immaterial. Should you change your opinion to better mesh with that of your would-be customers, please let us know. Many of us would like to buy your products if they become usable for our applications.
Assuming, of course, that "interpreted == slow" (and that bytecode language are "interpreted") - which is demonstrably false. The trick is to let the language do as much as possible. For example, writing mylist.sort() in Python is infinitely preferable to writing your own list sorting function (in Python) and using that instead. Similarly, using map() to apply a function to each value in a list is almost always going to be faster than manually iterating across it.
Again, let a language do what that language does best: implement itself. Learning what it's capable of and taking advantage of that will give you reasonable performance in all but extreme cases.
In reality, on the other hand, the reverse engineered drivers can serve as excellent documentation for how the same logic can be implemented in another OS, but that's about as close as it's likely to get.
On the other hand, my 2001-vintage PS2 still works flawlessly. See, that's the problem with anecdotal data - you can't extrapolate your own experience to everyone else.
Then world governments dictate that all WAPI-enabled router imports ship with an OpenVPN installer CD, and we all go the sane route of running trusted VPN software over untrusted open Wi-Fi connections.
True, but I could see them for being an attractive alternative to actually paying attention in "FLUFF101: Stuff Too Boring To Take Voluntarily" when you just want to learn the test well enough to get your A and get back to hacking Lisp.
I do. I never could get Reader Rabbit or Jumpstart Preschool running under Win2K, and I didn't want to buy (or pirate) XP when 98 runs my kids' games without problems.
Doesn't he have neighbors? People that fix his computer? Geeks who work at Pizza Hut spitting in his pizza? Pool cleaners who wizz in the hot tub when he's not looking? I have no desire to turn myself into a globally recognized pariah.
It's on the same LAN as my wife's relatively new iMac, a FreeBSD server, a Linux laptop, and an OpenBSD firewall. It's not that we're technically illiterate or poor, but that there's no legitimate need for us to upgrade the little gaming machine to something newer.
Have you discovered cperl-mode yet? It's like perl-mode but done right.
I think you read "as" as "since" instead of "while". I meant that you can be personally satisfied, but you'll still be facing the legal penalties.
Surprise, surprise: half the population has a below-average IQ. That's extremely relevant here because working through Kant's categorical imperative isn't something something for the faint of heart (or mind). If a researcher were naive enough to be surprised that people without the intellectual ability to also become researchers didn't operate in the same mental framework, then he should have his PhD revoked.
But you're right - you don't have to follow the law. You can still expect to be held accountable for not doing so, though, even if you disagree with it.
For a different example, a Randian Objectivist sees workplace regulations as inherently immoral. Not inconvient, not annoying, but immoral. And yet, a copy of "The Fountainhead" won't get him the hook for a failed OSHA inspection. You and he can both be morally satisfied that you followed your ethics as you pay off the fines.
You're only puzzled because you don't "get" open distributions like Debian. There are thousands of people working on it at any given time, from kernel packagers to people watching the bug tracker. Among that crowd, quite a few of them are passionate about legal issues and qualified to argue them. You're hearing those people. Everyone else is still going about their business.
This is in much the same way that discussions inside a company's legal department don't involve the programming department until a relevant decision is made. It's not like the programmers have to sit on their hands until the debates are finished, and probably don't even know they're happening at any given time.
So, what's your take on strongly typed, dynamic languages like Python (try adding 1 + "1" and see if it lets you)?
I think you missed the implied <sarcasm /> tag. I think it's idiotic too, but that was the advice we were given. I'm just relaying the message.
Yeah? When does that kick in? My wife's a surgeon, but I'm stuck driving a used Oldsmobile and lately riding a bike to work. We're not poor but neither do we live up to the "rich doctor" bullshit myth that you're buying into.
In reality, doctors spend the first 12 years after high school incurring enormous debts (you didn't think med school was free, did you?), then another decade in jobs that put food on the table but just cover a mortgage and student loans. Maybe we'll be wealthy some day - if only we can start turning a real personal profit by the time we're in our 40s.
So kiss my ass with the "doctors can afford it!" BS. Some can, but a whole lot of them can't. You might as well say "IT staffers have money", or "salesmen have money", or "accountants have money"; any of those would be just as accurate (and grossly overgeneralized).
That depends on your interpretation. By using names, you're theoretically announcing that Mrs. Smith has some unknown medical condition worthy of seeing a doctor. I think it's stupid, but I've been told exactly that by a compliance auditor.
Because her patients expect to be treated like humans and humans don't like being called by number. In her practice, patients tend to be retirement age or older (except for the occasional younger person with a broken toe, etc.). That population would not react well to being numerically processed.
She's much better off business-wise to upset the one person who's not used to the standard procedure at every doctor's office they've ever been to, than upsetting the 99.99% of her patients who like being recognized as a person with a name. You can argue about whether a practice should be run as a business, but if she drives off 50% of her clientele, then she won't be able to serve the other 50% at all.
He's got a point, though. That was an awful thing to have to endure, but that doesn't give Brady the right to lobby against my and your constitutional protections.
Never mind that we live in a small town where Mrs. Smith and Mr. Jones went to kindergarten together and come from families that have been here for 150 years. And forget that my wife is a podiatrist and that visiting her isn't inherently compromising (unlike, say, sitting in the lobby of a clinic for sexually transmitted diseases).
So, according to HIPAA, my wife is breaking the law each and every time she treats her patients like people instead of numbers. We haven't had a complaint yet and don't expect to, but could technically be busted for violating Mrs. Jones's privacy at any moment.